2016+AACE／ACE共识声明：门诊患者血糖监测

AACE/ACE Consensus Statement

AmERICAN AssOCIATION OF ClINICAl ENDOCRINOlOgIsTs AND AmERICAN COllEgE OF ENDOCRINOlOgy

2016 OUTPATIENT glUCOsE mONITORINg

CONsENsUs sTATEmENT

Timothy S. Bailey, MD, FACP, FACE, ECNU, Cochair1;

George Grunberger, MD, FACP, FACE,Cochair2;

Bruce W. Bode, MD, FACE3; Yehuda Handelsman, MD, FACP, FACE, FNLA4;

Irl B. Hirsch, MD5; Lois Jovanovi?, MD, MACE6;

Victor Lawrence Roberts, MD, MBA, FACP, FACE7;

David Rodbard, MD8; William V. Tamborlane, MD9; John Walsh, PA, CDTC10

From the 1Director, AMCR Institute Escondido, California Clinical Associate Professor, University of California, San Diego School of Medicine; 2Chairman, Grunberger Diabetes Institute; Clinical Professor, Internal Medicine and Molecular Medicine & Genetics, Wayne State University School of Medicine; Professor, Internal Medicine Oakland University William Beaumont School of Medicine Bloomfield Hills, Michigan; 3Atlanta Diabetes Associates; Associate Professor of Medicine, Emory University School of Medicine Atlanta, Georgia; 4Medical Director and Principal Investigator, Metabolic Institute of America; President, American College of Endocrinology Tarzana, California; 5Professor of Medicine, University of Washington School of Medicine Seattle, Washington; 6Physician Consultant, Sansum Diabetes Research Institute; Clinical Professor of Medicine, University of Southern California-Keck School of Medicine; Attending Physician-Santa Barbara County Health Care Services; Adjunct Professor, Biomolecular Science and Engineering and Chemical Engineering, University of California-Santa Barbara Santa Barbara, California; 7Professor of Internal Medicine, University of Central Florida College of Medicine Orlando, Florida; 8Chief Scientific Officer, Biomedical Informatics Consultants LLC Potomac, Maryland; 9Professor and Chief of Pediatric Endocrinology, Yale School of Medicine New Haven, Connecticut; 10Diabetes Clinical Specialist, AMCR Institute Escondido, California

Address correspondence to American Association of Clinical Endocrinologists, 245 Riverside Avenue, Suite 200, Jacksonville, FL 32202. E-mail: publications@https://www.wendangku.net/doc/bb10860039.html,. DOI: 10.4158/EP151124.CS

To purchase reprints of this article, please visit: https://www.wendangku.net/doc/bb10860039.html,/reprints. Copyright ? 2016 AACE.Abbreviations:

A1C = glycated hemoglobin; AGP = ambulatory glucose profile; ARD = absolute relative difference; BGM = blood glucose monitoring; CGM = continu-ous glucose monitoring; CMS = Centers for Medicare and Medicaid Services; CSII = continuous subcuta-neous insulin infusion; CV = coefficient of variation; DCCT = Diabetes Control and Complications Trial; DirecNet = Diabetes Research in Children Network; FDA = US Food & Drug Administration; GDM = gestational diabetes mellitus; GM = glucose monitor-ing; IDF = International Diabetes Federation; ISO = International Organization for Standardization; MARD = mean absolute relative difference; MDI = multiple daily injections; MedARD = median absolute relative difference; MNT = medical nutrition therapy; SAP = sensor-augmented pump; T1DM = type 1 diabetes mel-litus; T2DM = type 2 diabetes mellitus.

This document represents the official position of the American Association of Clinical Endocrinologists and American College of Endocrinology. Where there were no randomized controlled trials or specific U.S. FDA labeling for issues in clinical practice, the participating clinical experts utilized their judgment and experience. Every effort was made to achieve consensus among the committee members. Position statements are meant to provide guidance, but they are not to be considered prescriptive for any individual patient and cannot replace the judgment of a clinician.

ENDOCRINE PRACTICE Vol 22 No. 2 February 2016 231

232 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

INTRODUCTION

The measurement of glycemic status is a key element in the care of all persons with diabetes (1,2). Glucose mon-itoring (GM) enables clinicians to evaluate the efficacy of current therapy, make insulin and medication dose adjust-ments, ensure patients’ glucose levels are within therapeu-tic goal ranges, and monitor treatment safety. Both capillary blood glucose monitoring (BGM) and continuous glucose monitoring (CGM) with interstitial fluid sensors enable patients to better understand the impact of diet, exercise, illness, stress, and medications on glucose levels and to recognize and treat hypoglycemic and hyperglycemic epi-sodes. Likewise, both BGM and CGM have been shown to improve the efficacy and safety of diabetes therapy (3-12).

This document provides recommendations to clini-cians regarding the type and frequency of GM technology that should be employed in the management of patients with type 1 diabetes mellitus (T1DM: pediatric or adult), type 2 diabetes mellitus (T2DM), and pregnancy compli-cated by pre-existing diabetes or gestational diabetes mel-litus (GDM). In this document, we refer to GM technology that improves the lives of people with diabetes as “mean-ingful monitoring.” “The scope” of this document does not extend to the complexities of insulin adjustments based on the GM data obtained. Other pivotal reference documents can be consulted for this information (13,14). (Endocr Pract. 2016;22:231-261)

.

Additional aims of the document are to:

1. Provide a primer on GM accuracy

a. Describe various ways to characterize accu-

racy, such as mean absolute relative difference

(MARD)

b. Review GM accuracy guidelines from the

International Organization for Standardization

(ISO) and the US Food and Drug

Administration (FDA)

c. Discuss how device accuracy has the potential

to affect glucose control

2. Review measures of glycemic control (glucomet-

rics) such as the glycated hemoglobin (A1C) lab-

oratory measurement, change in average glucose

with time, percentage of time in target, hypogly-

cemic and hyperglycemic ranges, and glucose fre-

quency distribution. Graphical methods to display

glycemic data will also be presented.

History of GM in Diabetes

For several decades, urine glucose testing was the mainstay of diabetes monitoring (15). While patients could perform measurements at home and potentially adjust their therapy, the shortcomings of urine glucose testing were well recognized. Urine glucose correlated very poorly with blood glucose levels, provided no information about hypoglycemia, and gave negative results until the renal threshold for glucose excretion was exceeded. Therefore, urine glucose testing is presently of historical interest only.

The colorimetric Dextrostix? glucose test strip was developed in 1965. It was used for the first blood glu-cose meter in 1970 (15). Starting in the late 1970s, daily BGM gained wider acceptance as research data began to support the correlation and causation between poor glyce-mic control and diabetic complications (15-23). The “glu-cose hypothesis” was confirmed in the landmark Diabetes Control and Complications Trial (DCCT), the first long-term randomized prospective study to compare intensive (≥4x/day) self-GM coupled with an insulin titration algo-rithm versus standard therapy using once-daily GM and 1 to 2 daily insulin injections (24). Intensive therapy delayed the onset and slowed the progression of microvascular complications in patients with T1DM. Following the pub-lication of the DCCT results in 1993, the value of BGM in T1DM management became widely accepted, and its use gradually increased. It was clear that intensive insu-lin therapy and self-adjustment of insulin dosage in T1DM required frequent BGM (9,13,25-27).Subsequently, the effectiveness of BGM in GDM was demonstrated.

The value of BGM in T2DM has been controversial. As shown in Table 1, studies of BGM in T2DM have pre-sented mixed conclusions. Several have shown a clear ben-efit from frequent BGM (11,12,28-30). This has been par-ticularly evident for patients with T2DM who are receiving insulin therapy, especially involving multiple daily injec-tions (MDI), “basal-bolus” therapy, or insulin pump (con-tinuous subcutaneous insulin infusion) (31). Newer studies using a more structured testing approach have suggested benefit even for persons with diabetes not receiving insulin (9); these data support the need for patient education to ensure that each measured glucose leads to an action plan.

There is a common misperception that BGM is an expensive, complex undertaking with limited benefit, lead-ing some to assert that BGM is not warranted in patients with T2DM (32-35). The studies that appear to give nega-tive results in patients with T2DM have been criticized for serious experimental design flaws (28). Several stud-ies included rapid intensification of medication regimens following diagnosis, which may have obscured the effect of BGM. Additionally, many studies failed to couple GM to therapy adjustment, thus attenuating the benefit of the monitoring (28).

While BGM is a widely used and important compo-nent of T1DM therapy, it has drawbacks: patients’ moni-toring may be infrequent or intermittent, their reports may be inaccurate, and overnight glucose levels are seldom measured. Given these limitations, episodes of hypo- and hyperglycemia may be missed and not factored into treatment decisions (26,36). CGM offers the potential to revolutionize patient treatment by providing more fre-quent information that may allow a greater proportion

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 233

of patients to achieve target glucose and A1C levels with greater safety.

The first CGM device was approved in the United States in 1999. The MiniMed CGM System sampled glu-cose through a subcutaneously implanted sensor, recording glucose levels every 5 minutes over a period of 3 days. Initial versions of this technology did not provide glucose values in real time; data were downloaded and retrospec-tively evaluated by clinicians and used to make treatment adjustments (26). The first real-time CGM for prospective patient use was approved in 2001 (Glucowatch Biographer; Cygnus Inc, San Francisco, CA). The device used reverse iontophoresis to sample blood glucose, providing approxi-mately 36 measurements directly to patients over the 12-hour life of the sensor (37). It was withdrawn from the market due to skin site reactions, discomfort, limited accu-racy, and difficult setup and calibration procedures (38). Since then, CGM technology has improved dramatically in terms of accuracy, usability, and duration of use. The landmark Juvenile Diabetes Research Foundation (JDRF) Continuous Glucose Monitoring Study Group trial (6) established the role of CGM in T1DM, demonstrating sig-nificant A1C reductions in adults. The magnitude of ben-efit correlated positively with both wearing and interacting with the technology (4). In patients with lower baseline A1C, there were smaller reductions in A1C, but a reduc-tion in hypoglycemia (39). These benefits persisted for up to 12 months (40). Other unmasked parallel-group studies have confirmed significant reductions in A1C and a trend for reductions in severe hypoglycemia (3,4,41,42). A sum-mary of trial results for A1C and hypoglycemia reduction with CGM is shown in Figure 1.

CGM has the ability to provide alerts to actual or predicted episodes of hypo- and hyperglycemia. Further, all modern-day sensor devices display arrows reflecting the current slope of glucose versus time, which can assist clinical decision-making by the patient. However, CGM technology has drawbacks including expense; a need to frequently calibrate most devices; and some issues related to accuracy, comfort, convenience, and patient acceptance.

Current Status of GM

Previous publications from the American Association of Clinical Endocrinologists (AACE), Endocrine Society, and American Diabetes Association (ADA), provide sound general recommendations to guide diabetes therapy based on personal glucose records and laboratory values (1,2,28,43,44).No clinician caring for patients with dia-betes would dispute the value of employing some form of GM.

The Effective Health Care Program of the US Agency for Healthcare Research and Quality conducted compara-tive effectiveness research assessing GM methods and intensive insulin therapy methods. This included effective-ness studies comparing real-time CGM to BGM in adults, adolescents, and children with T1DM (45). While methods of GM did not affect patient quality of life, A1C was low-

ered by 0.3% in patients who used CGM compared with

234 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

patients who used BGM. This positive outcome for CGM was consistent for patients <18 years of age, supporting its use in adolescent patients and children. Unfortunately, because GM is a substantial cost driver in the management of patients with diabetes (28,46,47), governments and insurance companies have restricted coverage, payments, and reimbursement. However, improvements in A1C and accompanying reductions in hypoglycemia have been used to justify the cost of newer diabetes medications. To the extent that GM can also enable patients to achieve lower A1C values with less hypoglycemia, a similar and stronger case can be made for increasing access to GM (48), partic -ularly as costs come down and evidence continues to show benefit for both T1DM and T2DM. For patients who use insulin, CGM offers the distinct advantage of being able to securely maintain a more normal glucose range with less risk of hypoglycemia. As of the writing of this document, there remains no CGM coverage for elderly patients with T1DM, a population with frequent and severe hypoglyce-mia (49). Over the last 30 years, the FDA has approved many monitor models for use in GM. Since 2003, the FDA has required the accuracy of BGM devices to be within 20% of the true value at least 95% of the time (50). Certain monitors have shown substantially greater variability than allowed by FDA standards, leading to the recall of several brands of glucose meters and test strips in 2013 (51-54). The importance of GM accuracy and the emergence of stricter accuracy standards are discussed in greater detail in the “GM Accuracy and Precision” section later in this manuscript. In 2013, the US Centers for Medicare and Medicaid Services (CMS) implemented the controversial process of competitive bidding for BGM meters and test strips, with the intended goal of cost savings (55). This was one factor that led to a surge in the number and types of “generic” BGM meters. In some cases, when meters sourced from retail distribution channels were tested, the generic test-ing systems meters demonstrated dramatically inferior accuracy and precision compared to systems from major branded manufacturers (56-59). These generic meters showed sufficient performance data to obtain initial FDA clearance; however, they may not have maintained ade-quate performance over time, in part due to poor quality control leading to large between-lot variability in test strips. One proposed response has been to require postmarket sur-veillance of BGM products (60-62). The CMS competi-tive bidding process may have had other unintended con-sequences. A recent analysis of CMS data by the National Minority Quality Forum (NMQF) found that test areas in which competitive bidding was initially implemented had substantial disruptions in BGM supply acquisition com-pared to nontest markets (23% increase in partial acquisi -tion vs. 1.7% in nontest markets) (63). Within the test mar -kets, decreases in full acquisition (14.4%) and increases

in migration from full to partial acquisition (58.1%) were

Fig. 1. Glycated hemoglobin and hypoglycemia reductions in continuous glucose monitoring studies (189).

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 235

significant (P <.0001 for both) (64). Patients in these mar -kets had increased mortality and hospitalization rates and increased medical costs (63,64). Based on these results, the NMQF has called for the CMS to suspend competitive bidding until proper safety review and monitoring can be implemented (65). The purpose of the next section of this document, “GM Strategy and Rationale by Patient Profile,” is to provide concise and specific recommendations for clinicians on the type, frequency, and intensity of GM within the framework of specific patient profiles. The intent is to help clinicians counsel their patients to meaningfully monitor their glu-cose levels to optimize their diabetes care.GM STRATEGY AND RATIONALE BY PATIENT PROFILE T1DM

T1DM currently constitutes 5 to 10% of all people with diabetes globally (66,67). GM is one of the essential elements of effective T1DM management (68,69). The Type 1 Diabetes Exchange Clinic Registry (2013) found a systematic, statistically significant decrease in A1C levels in relation to increased frequency of daily BGM in chil-dren, adolescents, and adults (Fig. 2) (70).

Adult Patients With T1DM People with T1DM experience much greater glycemic variability than those with T2DM (71). This variability is associated with a higher risk of hypoglycemia (72). GM has a role in the early detection of hypoglycemia prior to overt symptoms.

BGM provides patients with important information regarding treatment efficacy (68,69). BGM can also facili -tate appropriate modifications to the therapeutic regimen, providing critical information that clinicians need to adjust dosage and/or timing of basal and bolus insulins, as well as reflecting the impact of food intake and physical activ -ity (2,68,73). Use of BGM is supported by clinical data: the DCCT, Epidemiology of Diabetes Interventions and Complications (EDIC), and many other clinical trials have clearly established the usefulness of BGM toward achiev-ing the goals of improved glycemic control and decreasing the risk of diabetes-related complications in T1DM (2,74). In all patients with T1DM, a rational and effective insulin regimen requires frequent GM. Frequent BGM is endorsed in all major clinical practice guidelines, including AACE, the ADA, the American Association of Diabetes Educators, the Joslin Diabetes Center, and the International Diabetes Federation (IDF) (2,28,68,73,75). Table 2 lists major organizations’ general recommendations for BGM timing and glucose goals in patients with T1DM. Current guidelines advise patients to check their blood glucose fre-quently; recommendations range from at least 4 to 6 to 10 or more times per day. All guidelines emphasize the need for individualization for each patient, with more or less fre-quent monitoring before meals, postprandially, at bedtime, before exercise, and when undertaking potentially hazard-ous tasks (e.g., driving) (2,68,69). Patients with T1DM should also monitor their blood glucose before driving and should not drive if their glucose level is <90 mg/dL (5.0 mmol/L).

Fig. 2. Association between blood glucose monitoring frequency and A1C in patients with T1DM (70). A1C = gly-cated hemoglobin; T1DM

= type 1 diabetes mellitus.

236 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

These guidelines also recommend the use of CGM, particularly for patients with a history of severe hypogly-cemia or hypoglycemia unawareness (1,2,44,68). Once again, the timing and frequency of monitoring must be individualized to meet specific patient needs (2,28). Table A1 in Appendix A of this document summarizes pivotal trials of CGM in adult and pediatric patients with T1DM. Pediatric Patients With T1DM

BGM remains a cornerstone for achieving optimal metabolic control in children, adolescents, and adults with T1DM (70). Frequent BGM, with a minimum of 4 blood glucose tests per day (premeal and at bedtime), should be the goal. In addition to these traditional 4 tests, many patients can gain a more robust picture of daily glucose trends by strategically adding additional tests, such as 2 hours after meals, overnight, and before and after exercise (76).

Optimal glycemic control of T1DM is particularly dif-ficult to achieve in pediatric patients. Food intake and activ-ity are unpredictable in very young patients, complicating parents’ efforts to regulate glucose levels. Additionally, many parents experience a “Scylla and Charybdis” situa-tion, where their fear that severe hypoglycemia will cause irreparable brain damage may lead to allowing a child’s glucose to “run high.” Data from the Type 1 Diabetes Exchange Clinic Registry indicate that children with elevated blood glucose and A1C levels are not protected against severe hypoglycemic events (77). Moreover, recent evidence from the Diabetes Research in Children Network (DirecNet) indicates that hyperglycemia is at least as det-rimental to normal brain development as hypoglycemia (78). In adolescents, the emotional fatigue of managing their diabetes often leads to a reduced frequency of BGM, missed insulin doses, and markedly elevated A1C levels. In older children and adolescents, the adverse effects of prolonged hyperglycemia on the cardiovascular system outweigh the potential harm from hypoglycemia (79), par-ticularly as treatment modalities and hypoglycemia man-agement strategies have improved (2).

Another special challenge of managing T1DM dur-ing childhood and adolescence is that insulin requirements change frequently. Simply measuring blood glucose and giving immediate correction doses are insufficient for long-term glycemic control in pediatric patients. Physicians, parents, and patients need to be instructed on how to rec-ognize trends that indicate the patient has outgrown their insulin dose(s) and learn to make longer-term regimen adjustments (80). Such pattern recognition requires main-taining and periodically reviewing an electronic or written log of blood glucose levels. Unfortunately, only a small proportion of physicians, patients, and families are down-loading data from glucose meters to appropriate computer programs; reviewing glucose meter data (including multi-ple graphs and statistics); and carefully making thoughtful, appropriate insulin dosage self-adjustments on a system-atic, periodic basis (81,82).

As in the case of BGM, CGM is only as beneficial as the patient’s desire and ability to use it. It is essential that all CGM users know the basics of sensor insertion, calibra-tion, and real-time data interpretation. To maintain a high frequency of use, patients and their parents require in-depth training with reinforcement, including periodic follow-up with clinicians and diabetes educators. The results of the JDRF CGM Study Group, using all the first-generation

CGM devices available at that time (2007), showed that

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 237

children, adolescents, and young adults (aged 8-24 years) who used the sensor almost every day benefitted clini-cally. Unfortunately, a much lower percentage of children and adolescents (34%) than adults (59%) performed daily CGM (83).

DirecNet studied the efficacy and safety of CGM in children <10 years of age. In a randomized clinical trial of 146 patients aged 4 to 9 years, CGM did not improve meta-bolic control. Despite a high degree of parental satisfaction with CGM, at the end of the 6-month study, only 41% of families reported daily CGM use (42). Similar results were reported by DirecNet in a nonrandomized, 6-month pilot study of 23 children <4 years of age (84). These studies were performed with older devices; the improved accuracy and ease of use of current devices might be better accepted. However, in a recent update of the state of the art of treat-ment of T1DM in the US, the T1D Exchange reported that <5% of youth <18 years old were currently utilizing a CGM device (85).

Combination of continuous subcutaneous insulin infusion and CGM (sensor-augmented pump)

The Sensor-Augmented Pump Therapy for A1C Reduction (STAR 3) Study (2012) examined a system that combines the use of a continuous subcutaneous insulin infusion (CSII) pump and a CGM system, termed sensor-augmented pump (SAP) therapy. In this 1-year study, chil-dren (aged 7-12) and adolescents (ages 13-18) with T1DM and baseline A1C ranging from 7.4 to 9.5% were random-ized to either SAP or MDI therapy. Overall, patients in the SAP group had significantly improved (P<.05) A1C values compared with the MDI group at all postbaseline visits (86). Furthermore, children and adolescents in the SAP group were consistently more likely to meet age-spe-cific A1C targets (88% and 57%, respectively) compared with those in the MDI group (51% and 13%, respectively) (86).Children and adolescents in the SAP group had lower area under the curve values than the MDI group, without increased risk of hypoglycemia, as well as improved glu-cose variability (86). STAR 3 was the first study to exam-ine the efficacy and safety of switching from conventional injections and BGM to 2 advanced technologies (CGM + CSII) nearly simultaneously; prior studies had only evalu-ated the impact of a single technology.

A SAP system with threshold suspend functionality was approved by the FDA in 2013 following consider-able experience in Europe. This device can suspend insu-lin delivery for up to 2 hours when the sensor glucose value reaches a predetermined lower threshold (87). The improved accuracy of CGM sensors and this threshold suspend (called “low glucose suspend” in Europe) may increase the performance and frequency of CGM use in pediatric patients. More recent studies have indicated the effectiveness of the predictive low glucose suspend system in children (88).

An international group of leading pediatric diabetolo-gists issued a 2012 consensus statement regarding the use of CGM in children (89). They recommended that CGM be considered for regular daily use in children and adolescents with T1DM who:

? Are performing frequent BGM

? Have experienced severe hypoglycemic episodes ? Have hypoglycemic unawareness, especially in young children

? Have nocturnal hypoglycemia

? Have wide glucose excursions, regardless of A1C ? Have suboptimal glycemic control, with A1C exceeding the target range

? Have A1C levels <7% and wish to maintain tar-get glycemic control while limiting hypoglycemia

risk

Accordingly, CGM is potentially applicable and desir-able in most children with diabetes. Recent enhancements have made it possible for parents and others to moni-tor glucose levels continuously via smartphones, wrist-watches, and computers. In 2015, the FDA approved mar-keting of 3 such systems: Dexcom Share (90), Dexcom G5 with Bluetooth (91), and MiniMed Connect (92). An open-source system (not FDA approved) called Nightscout was created (hacked together) by a group of people with diabetes and their families to allow remote monitoring by parents of children with diabetes (93). Other companies are likely to follow, as anecdotal reports suggest that par-ents and other caregivers find the technology invaluable when their children are away from home or participating in sports. Randomized controlled trial results evaluating these technologies are not available.

T2DM

Adult Patients with T2DM

BGM is an essential tool that should be accessible to all patients with T2DM, regardless of whether or not they are receiving insulin treatment (28). BGM is clearly ben-eficial for adult patients with T2DM because it provides immediate feedback regarding glycemic control (rather than requiring waiting, possibly months, for the next A1C measurement), and it assists with patient education, under-standing, and behaviors. Table A2 in Appendix A of this document summarizes pivotal trials of GM in adult patients with T2DM.

To ensure meaningful monitoring, use of BGM in patients with T2DM must be individualized by the physi-cian and healthcare team in partnership with the patient. The patient should be given specific guidelines including frequency and timing of testing and taught how to com-municate these results to the healthcare team. Methods for communication of glucose data are shown in Table 3. Two

238 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

of the goals for any BGM strategy are to empower patients to play a more active role in their diabetes management and to maximize the efficacy and safety of glucose-low-ering therapies, including lifestyle management (94). GM results are also a vital component of the data that should be presented to the diabetes care clinician at each medi-cal appointment, and potentially between visits, to assist in therapy titration.

Several randomized trials and literature reviews have called into question the clinical utility and cost-effective-ness of routine BGM in patients with T2DM who are not receiving insulin therapy (32,33,35,95,96). A key consid-eration is that BGM, used alone, does not lower blood glucose levels. To be useful, the information must be communicated to the healthcare team in an effective and timely manner and integrated into self-management plans. Several recent trials of structured BGM included specific instructions on testing frequency and timing, interpreting and communicating these results, and integrating results into self-management plans. These studies have shown improved glycemic control in patients with T2DM who do not receive insulin therapy (8,9,97,98).

General guidelines on the frequency and timing of testing based on specific patients’ diabetes therapy are pre-sented below and are outlined in Table 4.

GM in patients with T2DM on insulin therapy

If the patient is on intensive insulin therapy using prandial insulin combined with basal insulin, BGM should be performed when fasting, premeal, at bedtime, and peri-odically in the middle of the night. Such monitoring allows for appropriate adjustment of doses of premeal insulin, cor-rection boluses, and basal insulin.

If the patient is receiving only basal insulin, with or without other diabetes medications, BGM should be per-formed at minimum when fasting and also at bedtime to evaluate the impact of basal insulin on lowering overnight glycemic levels. If the decline in Bedtime to am(morn-ing) glucose (known as the BeAM factor) is >55 mg/dL (3.1 mmol/L), this suggests an excessive basal insulin dose (99), just as an overnight rise in glucose levels may indicate a need to increase basal insulin. Before titrating basal insulin to higher doses, consider improving the bed-time glucose by other means (e.g., with prandial insulin administered before dinner). This may prevent nocturnal hypoglycemia caused by excessive basal insulin and lead to improved overall glycemic control (31). If the patient is receiving basal insulin combined with 1 daily prandial or premixed insulin injection, BGM should be performed at minimum when fasting and before the prandial or pre-mixed insulin and periodically at other times (i.e., pre-meal, bedtime, 3 am, and possibly 2 hours postprandially). Insulin adjustments should be made to achieve acceptable glycemic targets.

GM in patients with T2DM on noninsulin therapies The IDF published a 2009 guideline specific to BGM in noninsulin-treated patients with T2DM (28). The IDF recommends that:

1. BGM should only be used when patients and/or

caregivers have the knowledge, skills, and will-

ingness to incorporate both BGM monitoring and

accompanying therapeutic adjustments into their

diabetes care plan.

2. BGM is only appropriate if protocols are indi-

vidualized to meet their patients’ educational/

behavioral/clinical requirements and have been

mutually agreed upon by the patient and clinician.

3. BGM should be considered both at the time

of diagnosis, to enhance patient education and

facilitate treatment initiation, and as part of

ongoing diabetes self-management education.

The goal is to help patients actively and effec-

tively participate in their treatment.

GM in patients with T2DM on noninsulin therapies associated with frequent or severe clinical problems related to hypoglycemia

Patients with T2DM receiving noninsulin agents asso-ciated with elevated hypoglycemia risk (specifically, sulfo-nylureas, and glinides) should perform BGM at least once daily (fasting) and periodically at other times to confirm the effectiveness of therapy and detect possible hypoglyce-

mia. Appropriate therapeutic adjustments should be made

gm Consensus statement, Endocr Pract.

2016;22(No. 2) 239

240 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

if patients are not at goal. Consideration should be given to altering therapy to employ 1 or more of the multiple classes that are not associated (or minimally associated) with increased risk of hypoglycemia (e.g., metformin, dipeptidyl peptidase-4 [DPP-4] inhibitors, sodium-glucose cotransporter-2 [SGLT-2] inhibitors, thiazolidinediones [TZDs], or glucagon-like peptide-1 [GLP-1] receptor agonists).

GM in patients with T2DM on noninsulin therapies not associated with hypoglycemia

Patients with T2DM receiving treatment regimens not typically associated with increased risk of hypoglycemia and who are not at goal should be instructed to perform structured testing (e.g., systematically before meals and at bedtime) at least weekly to adjust and confirm therapeutic effectiveness (9).Patients should be educated about when and how frequently to monitor glucose and should record the data in an organized logbook for subsequent review by a diabetes professional. Guidance for communication of glucose data is outlined in Table 3. After the A1C goal has been reached, and in the absence of evidence of hypogly-cemia, then less frequent monitoring may be necessary. GM in patients with T2DM on diet/lifestyle therapy only Daily BGM has not been shown to be effective in patients on diet/lifestyle therapy who are at low risk for hypoglycemia (28,33,35,94). However, structured testing may help patients improve their understanding of the effec-tiveness of medical nutrition therapy (MNT) and lifestyle management. Initial periodic testing at meals and bedtime provides feedback to the patient regarding the impact of various foods and physical activity on glycemic levels. After the goal A1C has been achieved, less frequent moni-toring may be needed.

Use of CGM in patients with T2DM

There are limited data on the use of real-time CGM in patients with T2DM, either masked for retrospective anal-ysis or unmasked for real-time use. Several studies have evaluated masked CGM, in which patients cannot see glu-cose values in real time, to help understand the progression from nondiabetes to prediabetes and T2DM (100). Other trials are ongoing to evaluate the potential use of masked CGM to guide both patients and clinicians regarding appropriate medication and lifestyle changes to improve glycemic control. Real-time CGM trials in T2DM patients are also ongoing, with several randomized controlled trials completed in recent years.

Vigersky et al compared real-time CGM (used for 8 of the initial 12 weeks of the study) to BGM 4 times a day in 100 patients with T2DM who were being treated with diet and exercise alone or with glucose-lowering therapies other than prandial insulin. At 12, 24, 38, and 52 weeks, respec-tively, this study found mean, unadjusted A1C decreases of 1.0%, 1.2%, 0.8%, and 0.8% in the CGM group com-pared with 0.5%, 0.5%, 0.5%, and 0.2% in the BGM group (P = .04). The reduction in A1C over the study period remained significantly greater in the CGM versus BGM group after adjusting for covariates (P<.0001). Patients who used CGM for at least 48 days showed the most improvement (P<.0001) (48).

A multicenter trial randomized 57 insulin-treated patients with T2DM to real-time CGM versus Internet-based BGM monitoring; results showed a greater reduction in A1C in the CGM group (1.31%) compared to the BGM group (0.83%), although the difference was not statistically significant (101). Additional randomized trials of CGM will be helpful in the evaluation of the benefits of CGM in T2DM.

Pregnancy Complicated by Diabetes

Approximately 8% of US pregnancies are complicated by either GDM or pre-existing T1DM or T2DM (102-104). In the early weeks of pregnancy, the excessively high maternal glucose levels of patients with poorly controlled or undiagnosed T1DM and T2DM are associated with an increased risk of miscarriage and congenital malformations (103,105). Hyperglycemia during the second and third tri-mesters results in fetal hyperinsulinemia that increases the risk of macrosomia and neonatal hypoglycemia (106,107). Maintenance of maternal glycemia as close to normal as possible through a program of BGM (or CGM), MNT, and insulin therapy offers the most effective protection against these complications (108).

The feasibility and efficacy of BGM in pregnancy complicated by diabetes were demonstrated in a seminal 1980 clinical trial that used BGM (8 measurements per day), MNT, and basal (neutral protamine Hagedorn) plus regular insulin in pregnant patients with T1DM (n = 10). All patients achieved normal mean plasma glucose and A1C levels, and the infants showed no signs of diabetes-related complications (109). Today, BGM is integral to the man-agement of diabetes in pregnancy (104).Real-time results enable individuals to make informed daily self-care deci-sions regarding diet, exercise, and insulin. Retrospective analysis of BGM data enables clinicians to develop indi-vidualized care plans (110), informing decisions related to insulin initiation and adjustment and the possible needs for interventions or hospitalization to improve inadequate self-monitoring (111).

CGM generates a detailed profile of glucose excur-sions that can be helpful when making decisions regarding self-care and treatment planning. Currently available CGM devices do not measure blood glucose levels <70 mg/dL (3.9 mmol/L) very accurately (112-114). Nevertheless, CGM can identify many episodes of hypo- and hypergly-cemia that would go undetected by BGM (108,115). CGM appears superior to BGM in this regard, but it remains to be seen whether CGM improves pregnancy outcomes. A 2013

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 241

trial comparing BGM alone to BGM combined with sev-eral 6-day periods of unmasked CGM in pregnant women with T1DM or T2DM (n = 154) found no differences in maternal A1C at term or in neonatal morbidity. Only 64% of the patients in that study were fully compliant with the CGM protocol, so potential benefits may have been missed. The most common reasons for noncompliance were device discomfort, sleep disturbances caused by alarms, and sen-sor inaccuracy (116).

The potential benefit of CGM for pregnant women with pre-existing diabetes is unclear based on currently available data. A prospective, randomized controlled trial performed in the United Kingdom assigned 71 pregnant females with T1DM or T2DM to prenatal care with or without CGM (117). While no maternal A1C differences were observed at baseline or throughout the first 2 trimes-ters, patients in the CGM group began to experience lower A1C levels between weeks 28 and 32, a difference that became statistically significant by weeks 32 to 36 (5.8% vs. 6.4%, P = .007). In contrast, a Danish trial that ran-domized 123 pregnant females with T1DM or T2DM to routine prenatal care alone or similar care plus CGM did not find any differences in outcomes between the 2 groups (118). Another randomized controlled trial of 340 Chinese females with GDM found that the use of CGM combined with standard care led to decreased A1C levels and less severe glycemic excursions compared to standard care alone (P<.001). Additionally, the use of CGM decreased the risk of pre-eclampsia and cesarean birth (P = .019 and P = .028, respectively) (119).

An ongoing study, the Continuous Glucose Monitoring in Women with Type 1 Diabetes in Pregnancy Trial (CONCEPTT, expected completion in late 2015), will attempt to determine if real-time CGM can safely improve glycemic control in patients with T1DM who are pregnant or planning pregnancy; this study will also assess infant outcomes (120).

CGM during pregnancy should be regarded as a teach-ing tool to evaluate peak postprandial blood glucose, fine-tune insulin dosing, and identify foods associated with blood glucose spikes (116). CGM can also be used as an adjunct to BGM to monitor nocturnal hypoglycemia and hyperglycemia, as well as the peak and duration of post-prandial hyperglycemia. A 2007 clinical trial of CGM in pregnancy reported that the additional information pro-vided by CGM altered clinical management decisions in 62% of cases (this trial did not evaluate patient outcomes) (121).

Table A3 in Appendix A of this document summarizes pivotal trials of BGM and CGM in patients with pregnancy complicated by diabetes. Blood glucose goals and recom-mended BGM patterns during and prior to pregnancy are summarized in Table 5.

Before attempting to become pregnant, females with pre-existing diabetes should maintain glycemic control as close to normal as possible for 3 to 6 months. Preprandial and fasting blood glucose should be maintained in the 60 to 90 mg/dL range, and postprandial glucose tested at 1-hour postmeal should be between 100 and 120 mg/dL (107).

The typical target fasting plasma glucose range during pregnancy complicated by diabetes is 55 to 90 mg/dL (3.1 to 5.0 mmol/L). This implies a heightened risk of hypo-glycemia. Accordingly, meter accuracy in the low blood glucose ranges is critically important in patients with preg-nancy complicated by diabetes. Hypoglycemia, in particu-lar asymptomatic hypoglycemia, is a key safety concern during pregnancy. Pregnant females with diabetes should monitor their blood glucose before driving and should not drive if their glucose level is <90 mg/dL (5.0 mmol/L). Likewise, they should always keep appropriate carbo-hydrate snacks with them in the car in case they become hypoglycemic.

GM ACCURACY, PRECISION,

AND DATA DISPLAY METRICS

Accuracy (the ability to obtain a true value without systemic bias) and precision (the ability to obtain highly reproducible results) have been steadily improving since the introduction of BGM in the 1970s. A 1986 ADA con-sensus conference, convened at a time when an estimated 1 million people with diabetes were using BGM, concluded that more than 50% of glucose meter measurements devi-

ated by more than 20% from a reference method. This was

242 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

attributed to both system and human factors. The ADA stated an aspirational goal in 1987 that 100% of BGM readings be within 10% of reference values (122). In 1993, a similar panel was convened and recommended that the analytic error not exceed 5% (123).Since it is only recently that any devices have even approached such performance, regulatory criteria for device approval have been more pragmatic, focusing on the hazards of incorrect readings (e.g., suboptimal treatment decisions, including improper adjustments in medication dosage, potentially increasing the frequency of both hypoglycemia and hyperglycemia) (58).

Accuracy, ergonomics, and ease of use of blood glu-cose meters have improved dramatically over time (124-126), and the accuracy of CGM is beginning to approach that of BGM devices (113,114,127-130). However, a clini-cally significant variation in accuracy and precision persists among currently marketed GM devices.Clinicians must be familiar with the clinical and laboratory standards used to characterize the accuracy and precision of the devices that they recommend in order to work safely and most effec-tively with patients using BGM or CGM systems.

Measures of BGM and CGM Accuracy There is a logical progression as to how one should interpret performance data with the objective of choosing the appropriate GM device for a particular patient. The fol-lowing presents such an approach.

(1) Bias. This refers to any systematic error in the measurements provided by the meter or sensor. This may be due to improper calibration, lack of calibration, or cali-bration with an inaccurate BGM. Bias may vary depending on the glucose levels being measured.

(2) Precision: Precision refers to the reproducibility of measurements, irrespective of whether they accurately measure the true value they are supposed to be measur-ing. Measurements may be highly reproducible but may be clustered around an erroneous value. We can measure the precision of a BGM or CGM by repeating measurements on the same blood sample or repeatedly measuring glucose using 2 or more CGM sensors simultaneously on the same subject. Even if the true value is not known, comparing the results for the multiple readings, we can derive a measure of precision.

For example, if 100 measurements gave a mean of 110 mg/dL with an SD of 5, the values would be very reproduc-ible with a percentage error of about 5%. However, if the true value were actually 100 mg/dL, then these measurements would be biased and would be significantly inaccurate.

(3) Arithmetic deviation: If the true value is 100 mg/ dL and the measured value is 110 mg/dL, then there is an arithmetic deviation of +10; similarly a value of 90 mg/dL would have an arithmetic deviation of ?10.

For example, if the result of the meter or CGM being evaluated is 85 mg/dL, and the true value is 100 mg/dL (as provided by a very precise and accurate laboratory method or by some other reference method), then the arithmetic deviation is ?15. These values can be calculated for each pair of true value and test-method value, and then aver-aged. The average should be extremely close to 0. One can then plot the arithmetic deviation versus the true value, to see if the average magnitude of the deviations varies sys-tematically with the true value (Fig. 3) Bias is defined as a systematic (built-in) error, which makes all measured val-ues wrong by a certain amount. As an overall estimate of bias, one can use the mean arithmetic deviation divided by the mean or average glucose level, expressed as a percent-age (131,132).

(4) Absolute deviation: The absolute deviation is the absolute value of the arithmetic deviation. In the cases above, the absolute deviations of the arithmetic deviations +10 and ?15 would be 10 and 15, respectively.

One should next examine the relationship of the abso-lute deviation and its average magnitude for various glu-cose ranges. There is almost always a systematic relation-ship between the absolute deviations and the true glucose level. If the true glucose level is not known, one can use the average value of multiple replicated measurements (Fig.

4).

(5) Absolute Relative Difference (ARD): Since an absolute deviation of 15 has a very different implication for a true value of 45 mg/dL compared with a true value of 400 mg/dL, it is common practice to express the absolute difference as a percentage of the true glucose. One can also plot ARD versus the true glucose levels as a continuous function (Fig. 5).

a. Mean Absolute Relative Difference (MARD):

When we calculate an absolute relative deviation

based on individual measurements using the meter

Fig. 3. Arithmetic deviations versus true glucose values (134). Relationship of deviations versus comparator glucose. The arith-metic (signed) deviations can vary in magnitude (bias) and in terms of their own variability depending on glucose level.

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 243

or CGM being evaluated (test method) as compared with a “true” laboratory-based method, there is a very large random sampling error. The mean absolute rela-

tive difference (MARD) is calculated as the average (mean) value of individual ARDs (133). To reduce the random sampling error in the measurement of ARD, it is desirable to calculate a MARD using a large num-

ber of paired test-comparator values for each specified narrow ranges of glucose (to achieve a 10% relative error in the MARD, it is necessary to have at least 500 data pairs).

MARD values have frequently been reported in the literature for the entire range of observed glucose levels (e.g., from 40 to 400 mg/dL). Since the ARD values differ systematically in the hypoglycemic, normoglycemic, and hyperglycemic ranges based on

a specific GM device’s performance, providing ARD

data for narrow glucose ranges gives important and useful performance information (134). MARD values for CGM can vary systematically by day of wear (e.g., day 1 vs. day 3 vs. day 7) (Fig. 6) (135,136). MARD also depends on rate-of-change of glucose.

b. Median ARD: Rather than using MARD, some

authors prefer to present results in terms of the median ARD.

One advantage of median ARD is that it is less influ-enced by outliers. However, it may be biased due to exclu-sion of the effects of outliers. Many studies have reported values for both MARD and the median ARD (frequently abbreviated as MedARD). MedARD is generally numeri-cally smaller than MARD. The ratio of the MedARD to MARD has been found to be approximately 0.8 empiri-cally for a variety of data in the literature. This is due to reduction in the influence of outliers, and the fact that the median is smaller than the mean for asymmetrical distri-butions such as ARD. It can be shown both empirically, using simulations, and theoretically, that the MedARD is approximately 0.8 MARD.

Table 6 summarizes the most commonly used terms that describe performance of glucose meters and sensors.

Understanding Clinical Standards for

Accuracy of Current BGMs and CGMs Error grids were the most popular early efforts to characterize the clinical significance of BGM device mea-surement errors. Regions of the grid are identified by let-ter designation, each reflecting the potential risk severity of incorrect treatment triggered by the measurement error (e.g., the device indicating hyperglycemia when someone is actually hypoglycemic). Clarke et al introduced the first error grid in 1987 (137). A variation of this grid was pre-sented by Parkes et al in 2000 (Fig. 7) to smooth the bound-aries of the grid regions. It incorporated the opinions of a greater number of expert clinicians (138). More recently, in 2014, a surveillance error grid with finer gradations in the categories for clinical error was introduced (139,140).

Device performance is typically reported as a percent-age of glucose values in zone A or zones A + B (higher percentages in zone A or zones A + B indicate better per-formance). However, there are no generally accepted tar-gets for clinical accuracy metrics such as percentage of observations within the various zones. These percentages may also depend on the range of blood glucose levels obtained. Error grids were a good tool to identify the fre-quency of egregious errors, but as meters have become more accurate, they are less useful for comparing device

accuracy.

Fig. 4. Absolute difference: average magnitude of absolute devia-

tions for various glucose levels (134). The absolute deviation of

the test method from the comparator shows large random sampling

variability. The magnitude of the absolute deviation and its own

variability depend on glucose level. The least-squares regression

line is shown.

Fig. 5. Absolute relative deviation as a continuous function of

true glucose (134).

244 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

Linear regression and correlation is another common way of expressing device accuracy (Fig. 8). Bland-Altman plots are used to illustrate the magnitude of errors depend-ing on the glucose level (Fig. 3) (134); these plots have been presented in a variety of formats. The vertical axis may show either arithmetic or relative error. The glucose levels shown on the horizontal scale may be the result of the comparator method or the average value of glucose measured by 2 methods subject to roughly comparable magnitude of error.

ISO Standards In 2003 the ISO criteria for glucose meters were intro -duced; the FDA adopted these the following year. Official meter approval standards from 2003 to 2014 are summa -rized in Table 7 (122,123,131,141-143). The 2003 ISO 15197 standard requires that 95% of the values be accurate within ±15 mg/dL (0.83 mmol/L) for glucose values <75 mg/dL (4.2 mmol/L) and within ±20% for glucose values ≥75 mg/dL (4.2 mmol/L). These were updated in 2013 (ISO 15197-2013) to require 95% of values to be accurate within ±15 mg/dL (0.83 mmol/L) for glucose values <100 mg/dL (5.55 mmol/L) and within ±15% for glucose values ≥100 mg/dL (5.55 mmol/L) (131,142). On January 7, 2014, the FDA released draft guidance for BGM accuracy that would require far more accuracy and precision from BGMs (143). The draft proposes that there be smaller errors in the hypoglycemic range and fewer outliers, allowing only 5% of measurements to have an error larger than ±15% and 1% of measurements to have an error greater ±20% above or below the reference value, rather than the 5% permitted under the 2003 ISO Guidelines. Further, the FDA was considering requesting that the experiment test be repeated 3 times, and the device would need to pass all 3 tests. This would make the test -ing more rigorous and conservative. If devices are tested by trained technicians, one would expect greater accuracy than if they were tested by untrained lay-persons such as patients and family caregivers. There is a suggestion that testing performed by nontrained people under “real-world” conditions might become required (144). Not all BGMs that receive FDA approval provide the same degree of accuracy. Several published studies have compared BGM brands and models by name during head-to-head testing (56,57,136,145-148). For clinicians and consumers, MARD provides an excellent measure of accu-racy and precision when evaluating a BGM (134). It has also been recommended that bias and coefficient of varia -tion (%CV) should be reported (one can show mathemati -cally and by simulations that there is a direct relationship between MARD and %CV: MARD is approximately 0.8 %CV) (132). The degree of BGM accuracy that is desired and required is likely to depend on the clinical needs of

individual patients. There is a growing consensus among

Fig. 6. CGM MARD values displayed by day of wear (135). Box plots for MARD on successive study days. Displayed are mean (diamonds), median (horizontal lines within boxes), 25th and 75th percentiles (lower and upper edge of the boxes), and minimum and maximum values (antennae). CGM = continuous glucose monitoring; MARD = mean absolute relative difference.

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 245

endocrinologists and other clinicians that the accuracy and precision performance characteristics of each BGM and CGM device should be made available both to the patient and physician, to properly match a BGM device to the appropriate individual or clinical setting (149).

How Much Accuracy Is Needed?

Research on the impact of GM inaccuracy on health outcomes is limited; however, computer modeling can separate the impact of GM errors on glucose outcomes from those due to other factors. Modeling studies indicate that patients receiving bolus insulin therapy face increased risk of hypoglycemia even when using GM devices that achieve current standards (140,150-153).

One study used 100 simulated adults with T1DM to run 16,000 virtual trials applying varying levels of simulated BGM error (5%, 10%, 15%, and 20% deviation from true blood glucose values). Results showed that glycemic con-trol deteriorated with each increase in BGM error. Failure to detect hypoglycemic episodes, hypoglycemia risk, gly-cemic variability, and A1C increased as BGM error level increased (150). In another study, Schnell and colleagues reported that improvements in BGM accuracy (reducing error from ±20 to ±5%) would be expected to result in a 10% reduction in severe hypoglycemia, a 0.4% reduction in A1C levels, and a 0.5% relative reduction in myocardial infarction. This study (2012 data) estimated an annual cost savings from this kind of improvement in BGM accuracy of €9.4 million for patients with T1DM and €55.5 million for insulin-treated patients with T2DM for Germany alone (151).

Another study of 100 simulated cases being treated with intravenous insulin therapy in an intensive care set-

ting found that increases in either BGM imprecision

246 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

(measured as %CV) or bias, tested separately (with 1 or the other variable set to 0), increased glucose variability and the frequency of hypoglycemia and hyperglycemia (154). BGMs with a %CV ≤6.5% and bias ≤5% rarely lead to major (2-step or greater) errors in insulin dosing. This degree of accuracy would ensure that the rate of any insu-lin dosing errors would be <5% (155). Table 8 summarizes clinical situations where increased accuracy may be of par-ticular benefit.

What Impacts Accuracy?

Manufacturing defects and test-strip lot-to-lot variations directly impact accuracy and introduce bias (156,157). Bias is typically measured in the hypoglycemic range, target range, and hyperglycemic range. One study of test-strip accuracy compared 7 meters and tested 3 test-strip lots for each range and found that lot-to-lot variations were as high as 11% using the same meter (158). Another study found that the difference in bias between widely used BGM devices was as high as 4.8% (159). Underfilling the test strip can introduce errors >20% in some BGMs. In another study, only 5 of 31 glucose meters were able to maintain 100% accuracy (either giving the correct read -ing or rejecting the reading appropriately) when test strips were deliberately underfilled (160).

Although many meters have been approved for alter-nate site testing (e.g., sampling from the palm, upper arm, forearm, thigh or calf, rather than the fingertip), this prac -tice can generate inaccurate results, particularly when glu-cose levels are changing rapidly such as after meals or after exercise, when the patient is ill or under stress, or shortly after insulin administration (68).

BGM testing methods are predominately based on the glucose oxidase or glucose-1-dehydrogenase enzyme. Any factor that interferes or impacts these enzymes or the BGM itself can degrade overall accuracy. Variation can be due to issues such as competing blood substrates (e.g., maltose, vitamin C) (161,162), environmental issues (e.g., cold tem-perature, high altitude with reduced oxygen pressure), and factors related to individual patients. Reduced accuracy and precision have been observed in tests performed by patients and other lay users compared with highly trained, experi-enced health professionals (163). GM accuracy is just one of many factors influencing the quality of subsequent glyce -mic control achieved. Contaminants on the skin from food sources (fruits, juices, sodas, milk) and even hand lotions can artificially raise capillary blood glucose levels and potentially lead to an overdose of insulin with subsequent hypoglycemia. Acetaminophen is well-known to result in spurious values in CGM systems (15,44,56,164,165). Physical compression of the CGM sensor during sleep can result in seriously low glucose readings.

How to Communicate Device Accuracy Data

It would be highly desirable to be able to label each GM device and its test strips or sensors with their per-formance characteristics, and methods for labeling have been contemplated for several years (166,167). In a recent guidance document (143), the FDA suggests a simple sys -tem that shows the percentage of a BGM glucose values expected to fall within 5%, 10%, 15%, and 20% of the reference values (Fig. 9) (143). This allows clinicians and patients more insight into the accuracy of a particular GM

device so they can make an informed choice.

Fig. 7.

Parkes error grid (138).

Fig. 8. Linear regression relationship between observed and com-parator glucose (134).

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 247

BGM Accuracy Is Necessary but not Sufficient to Improve Quality of Glycemic Control

As measurement tools, BGMs and CGMs generate data used to make treatment decisions and adjust diabe-tes medication doses. The aptitude of patients and clini-cians with regard to data analysis and interpretation var-ies widely. Accordingly, the methods of data display and reporting are critically important. Older BGMs displayed a single value without context. In contrast, many current BGMs report weekly or monthly averages for glucose and may also highlight patterns in glycemic variability (e.g., consistently high or low values at a particular time of day or in relationship to a specified meal). Similarly, current CGM devices have on-screen analysis capabili-ties that display glucose trend lines over time, with arrows reflecting the magnitude of the current rate-of-change of glucose. These features provide additional information and help give context to raw glucose numbers. However, many users will require guidance to effectively use these infor-mative features.

Clinicians should also consider the ease and speed of BGM downloading to ensure that the end user will be able to identify glucose patterns and that clinical interventions will be properly implemented. Currently, each device has proprietary software that displays data in widely differing formats, making clinical interpretation difficult. To accom-modate their patients, clinicians need to master multiple software products. Although no current software down-loads every device, several companies and organizations are attempting to develop standardized methods to down-load and display data from nearly every type of BGM, CGM, insulin pump, and other health devices (e.g., activity monitors).

To correctly gauge the timing of hypoglycemic and hyperglycemic events, the clock setting in the BGM must be accurate (168). BGM clock settings should be clearly visible and easy to adjust and should remain accurate when a battery is changed or temporarily removed. Clocks in the meter, CGM, and insulin pump (if utilized) should be syn-chronized (automatically if possible), with accommodation for travel across time zones. Ideally, all glucose and related data should be integrated with an electronic health record.

It has been proposed that the ongoing routine quality assurance verification currently being performed by manu-facturers to ensure the accuracy and precision of subse-quent lots of test strips should be confirmed by indepen-dent laboratories using a standardized methodology (146). In support of this concept,Freckmann and colleagues reviewed the accuracy of 27 meters previously approved in Europe under the 2003 ISO 15197 standard (±20% for glu-cose levels >75 mg/dL and ±15 mg/dL for glucose levels ≤75 mg/dL). In postapproval testing, more than 40% of the meters failed to meet the standard by which they had previ-

ously received approval (58). When people with diabetes

248 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

Fig. 9. Sample label information for meter and test-strip boxes (From the US Food and Drug Administration Guidance Document) (143).

gm Consensus statement, Endocr Pract. 2016;22(No. 2) 249

performed the testing, fully one-third of meters failed to meet the 2003 ISO 15197 standards (169). A recent study showed that only 12 (44.4%) of 27 available BGMs met the most recent 2013 ISO 15197 standard. Only 13 of 27 (48.1%) BGMs gave adequately accurate results in the hypoglycemic range, while 19 (70.3%) had sufficient accu-racy for glucose levels >250 mg/dL (13.9 mmol/L) (170). Unfortunately, one cannot assume that FDA approval implies that a BGM will continue to meet FDA accuracy requirements for subsequent batches of test strips.

Glucometrics, Downloading,

and Interpretation of GM Data

The analysis and display of glucose data is termed “glucometrics” (171). It can describe the average value, distribution of glucose, glucose variability, patterns during the day and night, effects of days of the week, and long-term trends. The availability of GM devices with electronic memory and the ability to download these data has fueled the rapidly growing science of glucometrics. Retrospective analysis of glucose levels, both overall and at specific times (e.g., after major meals or on selected days of the week), can provide insights into how factors such as medications, diet, stress, and activity contribute to diabetes control and how those factors should be addressed or adjusted (82,172). Communication of glucometric data to the healthcare team is key; communication methods between patient and clini-cian are presented in Table 3.

Which glucometrics parameters are best? Approaches vary in complexity but usually generate similar types of information (171,173).With enough information, it becomes possible to evaluate whether the A1C level, still the gold standard, is consistent with the patient’s average blood glucose (174).

Table 9 summarizes high-level, clinically relevant information that can be obtained from BGM or CGM data. Either the mean or median can be used to charac-terize the average glucose level. Since the SD of glucose is fairly highly correlated with the mean glucose, %CV is usually the best single simple method to characterize vari-ability (26,37,175-178). As an approximation, SD tends to be higher in patients with higher mean glucose values. While mean, median, and %CV metrics describe overall glycemia, several additional methods have been developed to describe actionable patterns to help clinicians optimize diabetes therapy. In a graphical presentation, the “stan-dard day,” “modal day,” (179,180) or ambulatory glucose profile (AGP) displays individual glucose measurements (pooled over multiple days) by time of day on a single 24-hour scale (Table 9; image 1A; image 1B.; image 2A.). This graph indicates both the glucose values and the times of day when people have been monitoring their glucose levels, facilitating the detection of any consistent patterns in glucose excursions and providing an assessment of the adequacy of GM. The “Standard Day” is simple in prin-ciple but can be difficult to interpret in view of the large amount of scatter observed in glucose data obtained over several days.

AGP

The AGP was introduced by Mazze et al (1987) for BGM and subsequently applied with further enhancements (display of the smoothed curves for the 10th and 90th per-centiles) to CGM data by Mazze (2008) and Bergenstal and colleagues (2013). The AGP provides an excellent starting point for a standardized computerized display of BGM and/or CGM data by time of day (173,178,179). To gener-ate the AGP, an individual’s blood glucose levels are mea-sured via CGM or BGM with all glucose data pooled and analyzed as if it had been collected during a single 24-hour period. The result is a standardized software report that can be displayed graphically. Examples of graphic AGP dis-plays for patients with normal glucose tolerance, T1DM, and T2DM are shown in Table 9 (images 2A-C) (173,181). AGP has been proposed as a standardized method for glu-cose reporting and analysis (173,178,181). One can also examine these 24-hour patterns in glucose by day of the week (180).It is customary to report a number of statistics to accompany the graphical display of the AGP (173).

Several additional graphic displays of data related to changes in glucose over time, time within different glucose ranges, glucose profile, etc. are shown in Table 9. Some are simplistic (e.g., pie graphs or simple bar charts display-ing percentages of glucose values above, below, and within the target range). Others are slightly more complex (e.g., box plots [a methodology introduced by Tukey as part of his approach to Exploratory Data Analysis that makes no assumptions about the nature of the underlying distribution of glucose values and was introduced into glucometrics by Rodbard (180,182)], scattergrams, stacked bar charts, and histograms). Their purpose is to help the clinician identify and prioritize clinical problems and then educate and moti-vate the patient to achieve improved glycemic control. Recommendations

Health professionals should educate patients regard-ing the interpretation and use of GM data to help modify patient behaviors, enhance their ability to self-adjust ther-apy, and help them decide when to seek medical assistance.

To assess glucometrics, first examine the overall sta-tistics (mean, SD, %CV); distribution of glucose values (e.g., stacked bar charts); and glucose by date, time of day, in relationship to meals, and by day of the week. This docu-ment provides several examples for each of these types of analyses. Usually, the most helpful are graphs of glucose by date, the AGP by time of day, stacked bar charts in relation-ship to time of day, and stacked bar charts and “box plots” for glucose in relation to meals and by day of the week.

250 gm Consensus statement, Endocr Pract. 2016;22(No. 2)

(Continued next page)

危重患者血糖的监测及控制习题及答案

危重患者血糖的监测及控制姓名：科室：得分： 一、选择题： 【A型题】 1、下列选项中不是高血糖发病机制的是：（ D ） A.胰岛素抵抗 B.应激 C.严重感染 D.肾上腺危像 E.胰岛B细胞功能丧失 2、胰岛素注射优先选择的部位是：（ A ） A.腹部 B.大腿内侧 C.臀部 D.手臂 E.静脉给药 3、以下哪种胰岛素属于速效胰岛素（ B ） A.诺和灵 R B.诺和锐 C.诺和灵N D.甘精胰岛素 E.甘舒霖N 4、下列哪项指标反映糖尿病患者近2-3月血糖控制不佳（ D ） A.尿糖阳性 B.血糖＞30mmol/L C.尿酮体强阳性 D.糖化血红蛋白A1为% E.基础血浆C-肽水平低于400pmol/L 【B型题】

A.阴离子间隙增大 B.血浆乳酸5mmol/L C.尿糖阴性 D.空腹血糖L E.血钠155mmol/L 5、高渗性昏迷（ E ） 6、糖尿病酮症酸中毒（ A ） 7、低血糖昏迷（ C ） 8、糖尿病（ D ） 9、乳酸性酸中毒（ B ） 【多选题】 10、下列关于重症患者血糖控制中正确的说法是：（ ABCDE ） A.解除应激因素 B.合理输注葡萄糖 C.静脉使用胰岛素 D.根据血糖监测情况调整胰岛素用量 E.一旦出现低血糖，静脉给予50%葡萄糖输入 二、简答题： 1、高血糖的危害： 答： 1.高血糖可以降低机体的免疫功能。2.高血糖也可降低补体的活性, 糖通过补体进行糖化作用, 和微生物竞争与补体的结合, 抑制调理作用。3.高血糖是加重机体能量代谢障碍的重要影响因素。高度应激状态下的高能耗、高氧耗和高分解代谢更加重了机体能量储备的耗竭, 最终可以导致细胞水肿、溶解和器官功能衰竭。 2、患者发生高血糖的机制： 答：高血糖是由于激素和细胞因子水平增加引起的。1)升糖激素在应激状态下, 下丘脑-垂体-肾上腺轴(HPA)过度兴奋,促分解激素, 如糖皮质激素、胰升糖素、生长激素、儿茶酚胺等分泌增多, 而胰岛素分泌相对减少, 胰升糖素/胰岛素比例失调, 从而使糖异生增加, 肝糖原和肌糖原分解增加, 糖的生成速率明显增加。2)细胞因子在严重应激状态下, 来自不同组织的多种细胞因子对高血糖的产生也具有十分重要的作用。已知细胞因TNF-α主要通过调节胰岛素受体后信号的传导, 造成肝脏和骨骼肌对胰岛素耐受而使血糖增高。

快速血糖监测技术

中铁二局二处医院 快速血糖监测技术 血糖：血液中所含的葡萄糖 血糖测定：判断糖尿病病情和控制血糖情况的主要指标 血糖仪监测血糖的原理：通过测量血液中的葡萄糖与试纸中的葡萄糖氧化酶反应产生的电流量测量血糖。 采集末梢血：滴血式和吸血式 优点：较生化仪测量更快（30秒内）用血量更少（5微升以下） 血糖仪的特点：方便、快速、安全。 只适合日常检测，不能作为准确诊断糖尿病的工具，应进行及时校准和常规质控。 血糖值： 正常人：空腹：3.9～6.1mmol 餐后2小时：≤7.8mmol/l 糖尿病：空腹：≥7mmol/L 餐后2小时：≥11.1mmol/L 血糖危急值： 成人血糖低于3.9mmol/L 高于16.7mmol/L 择期手术者应把空腹血糖控制在7.8mmol／L以下，餐后两小时血糖控制在11.1mmol／L 急诊手术者随机血糖应控制在13.9mmol/L以下。 如空腹血糖＞10mmol/L、随机血糖＞13.9mmol/L或糖化血红蛋白（HbA1c）水平＞9%，应推迟非急诊手术。 眼科手术等一些精细手术对血糖要求更加严格，应控制在 5.8～

6.7mmol/L。 血糖异常： 1.生理性高血糖可见于摄入高糖饮食或注射葡萄糖后，或精神紧张、交感神经兴奋，肾上腺分泌增加时。 生理性低血糖饥饿或剧烈运动、注射胰岛素或口服降血糖药过量。 2. 病理性高血糖 ⑴糖尿病： ⑵对抗胰岛素的激素分泌过多：如甲亢、肾上腺皮质功能及髓质功能亢进、腺垂体功能亢进、胰岛α-细胞瘤等。 ⑶颅内压增高：颅内压增高（如颅外伤、颅内出血、脑膜炎等）刺激血糖中枢，出现高血糖。 ⑷脱水引起的高血糖：如呕吐、腹泻和高热等也可使血糖轻度增高。 4. 病理性低血糖 ⑴胰岛素分泌过多：由胰岛β细胞增生或胰岛β细胞瘤等引起。 ⑵对抗胰岛素的激素分泌不足：如腺垂体功能减退、肾上腺皮质功能减退和甲状腺功能减退等。 ⑶严重肝病患者：肝贮存糖原及糖异生功能低下，不能有效调节血糖。低血糖临床表现可归纳为以下两个方面。 1.自主（交感）神经过度兴奋的表现 低血糖发作时由于交感神经和肾上腺髓质释放肾上腺素、去甲肾上腺素等，临床表现为出汗、饥饿、心慌、颤抖、面色苍白等。 2.脑功能障碍的表现

血糖监测登记表

血糖监测登记表

如何规范监测血糖？ 血糖监测频率及时间点均需根据患者实际情况进行个体化决定： ★毛细血管血糖监测频率与时间点 （1）生活方式干预以控制糖尿病的患者：据需要有目的地进行血糖监测来了解饮食运动对血糖影响并予以调整。 （2）服口服降糖药的糖尿病患者：每周 2-4 次空腹或餐后 2 h 血糖，或就诊前一周连续监测 3 天。每天监测早餐前后、午餐前后、晚餐前后、睡前这 7 点血糖。 （3）使用胰岛素强化治疗的糖尿病患者：需每周至少3天，每天监测5～7个时间点血糖，包括早餐前后、午餐前后、晚餐前后、睡前血糖，对于怀疑有黎明现象或苏木杰反应的患者应增加监测0AM和3AM的血糖，依据血糖检测结果调整胰岛素剂量。 ●基础胰岛素使用者：监测空腹血糖并据之调整睡前胰岛素剂量； ●预混胰岛素使用者：监测空腹+晚餐前血糖，据空腹血糖调整晚餐前胰岛素剂量、据晚餐前血糖调整早餐前胰岛素剂量； ●餐时胰岛素使用者：监测餐后或餐前血糖，据餐后血糖和下一餐前血糖调整上一餐前胰岛素剂量。 （4）特殊人群的糖尿病患者：遵循以上基本原则并据情况可实行较宽松血糖控制标准，包括：围手术期患者、低血糖高危人群、危重症患者、老年患者、1 型糖尿病、妊娠期糖尿病等。 ★因血糖控制非常差或病情危重而住院治疗应每天监测4-7次血糖或根据治疗需要监测血糖，直到血糖控制。注意因操作不当、血量不足、局部挤压、更换试纸批号校正码未更换、试纸保存不当等影响检查结果。 ★糖化血红蛋白HbA1c检测频率：与毛细血管血糖反映的时间窗不同，糖化血红蛋白反映过去2～3个月血糖控制的平均水平，它不受偶尔一次血糖升高或降低的影响，因此对糖化血红蛋白进行测定，可以比较全面地了解过去一段时间的血糖控制水平，是临床上评估长期血糖控制状况的金标准，也是临床决定是否需要调整治疗的重要依据。建议在治疗之初至少每3个月检测1次，达到治疗目标可每6个月检测1次。

快速血糖监测技术考核评分标准

快速血糖监测技术考核评分标准 科室：姓名：得分： 项目总分技术操作要求标 准 扣分及原因 仪表 5 仪表端庄、服装整洁 5 核对 2 双人查对医嘱 2 评估15 1. 评估患者：姓名、年龄、诊断、病情、意识状 态、合作程度及穿刺手指末梢循环及皮肤情况（有 无疤痕、炎症、硬节等） 2.了解患者进食、水情况，是否符合空腹或餐后2 小时血糖测定要求 3.指导或协助患者用温水或肥皂水清洗双手 4. 评估患者周围环境 5.告知患者：对清醒患者，要告知血糖监测的、目 的、方法及配合，问二便。 5 5 1 1 3 操作前10 1.洗手，戴口罩 2.用物准备：治疗盘内置75%酒精、棉签、棉签、 血糖仪、同型号血糖试纸、采血针、执行单 3.检查血糖仪性能是否完好，仪器表面是否干净、 清洁，血糖试纸是否在有效期内 2 5 3 操作中43 1.查对患者及检测项目 2.协助患者取舒适体位 3.从患者手腕向指尖按摩2-3次促进指尖血液循 环 4.75%酒精消毒指尖，待干 5.从试纸瓶内取出试纸，并快速盖紧瓶盖。查看试 纸表面有无受潮或污染，禁用手触摸试纸两端 6.沿箭头方向将试纸插入到血糖仪试纸插口中，血 糖仪自动开机 7.确认血糖仪显示的密码和试纸筒标签上的密码 一致 8.显示屏上显示出线试纸和闪烁的血滴符号 9.再次核对患者及检测项目 10.用拇指和食指固定采血的指间关节，用采血针 在指尖任一侧刺破皮肤采血（避开指腹神经末梢 丰富部位，减轻疼痛） 11.弃去第一滴血液，用第二滴血液进行测试 12.将血滴接触试纸黄色窗口的前沿位置，血液自 动吸入黄色窗内，直至填满整个黄色窗口。看到 有图标闪烁是，代表试纸已经有足够的血样，如 2 1 2 2 2 2 2 2 2 5 2 10

危重患者的血糖控制措施

危重患者的血糖控制措施 陆菊明解放军总医院内分泌科 2010-7-12 点击：2016 机体处于应激状态时，体内多种升高血糖的激素如肾上腺素、糖皮质激素、胰高糖素、生长激素等水平明显增高，导致应激性高血糖，这在糖尿病患者中尤为明显。内外科的危重患者，不管是否有糖尿病史，此时常存在高血糖症和胰岛素抵抗。现结合近年来有关住院危重患者胰岛素强化治疗的临床试验，探讨此类患者血糖控制目标和治疗措施。 血糖控制标准受关注 2001年比利时van den Berghe等进行的Leuven研究发现，严格控制外科ICU危重患者血糖在正常范围（4.4～6.1mmol/L）可降低死亡率，该结果发表之后引起了广泛关注。但在van den Berghe等随后的一项研究内科ICU危重患者结果却显示，强化治疗组的主要转归指标住院死亡率（37.3%）与常规治疗组（40.0%）相比无显著降低，但强化治疗可降低并发症发生率、缩短机械通气时间及住院时间。 许多国家的糖尿病诊治指南都纳入了上述研究结果。2009年美国糖尿病协会（ADA)的指南对于危重患者血糖控制要求尽量接近6.1mmol/L。2007版中国2型糖尿病防治指南也指出，术后需要重症监护或机械通气的高血糖患者（＞6.1mmol/L），通过持续静脉输注胰岛素而尽可能使血糖控制在4.5～6.0mmol/L范围内可改善预后；但也提到较为保守的血糖目标（6.0～10.0mmol/L）在某些情况下更为合适。 严格控制血糖有风险 一些类似研究（如GLUCONTROL等）发现危重患者进行强化降糖治疗并未获益，VISEP 研究甚至因强化控制血糖组中低血糖发生率过高而被迫提前中止。2008年发表在《JAMA》的一项荟萃分析也得出了不同的结论，即对于危重患者，强化控制血糖与住院死亡率降低无关，且可增加低血糖症的发生风险，但能降低败血症的发生风险。 2009年3月《新英格兰医学杂志》发表了NICE-SUGAR试验结果，对Leuven的研究发

血糖监测系统测试题+答案

体外诊断检验系统 自测用血糖监测系统通用技术条件培训测试题 一、填空：（共82分，一空2分，共41空，41*2=82分） 1.GB/T 19634-2005本标准规定了自测用血糖监测系统的术语和定义、要求、试验方法、标签和使用说明、包装、运输和贮存。 2. GB/T 19634-2005本标准适用于体外监测人体毛细血管全血和/或静脉全血中葡萄糖浓度的自测监测系统，通常包括便携式血糖仪、一次性试条和质控物质。 3.当“准确度”这一术语应用于一批检测结果时，包括了随机误差部分和一个共有的系统误差或偏差部分。 4.血糖监测系统包括一个便携仪器和与其适配的试条及质控物质，用于体外监测血糖浓度的检测系统。 5.系统准确度可以表示为将评价系统所得结果与其参考值之间差异的95%包括在内的区间。参考值是指一个可以溯源至较高级水平参考测量程序所赋的测量程序所赋的值。 6.精密度的程度是用统计学方法得到的测量不精密度的数字形式表示的。如标准差（SD）和变异系数（CV），都与精密度成负相关。 7.制造商提供的自测用血糖监测系统信息手册，内容包括试剂的安全和正确使用或仪器的安全和正确操作、保养和基本的解决故障的方法。 8.质控物质是制造商用于一个体外诊断医疗器械的性能特征的物质、材料或无物品。质控范围是质控物质测得的结果在统计学上可接受的数值区间，测量结果95%应在质控范围之内。 9.血糖试纸通过化学或电化学反应产生信号，用以检测存在于样品中的葡萄糖的浓度。 10.血糖仪外观应光洁无毛刺；正面的加样区洁净无污点；质控物质外观应均匀一致。 11.血糖仪和血糖试纸重复测量结果的精密度应符合在<5.5mmol/L ，SD<0.42mmol/L ;≧5.5mmol/L ，CV<7.5%.准确度测量结果应符合≦4.2mmol/L ，允许偏差±0.83mmol/L ;>4.2mmol/L，不超过±20%. 12.不同批血糖试纸之间测量值的差异称为批间差，应不大于15%。 13.重复性实验用血糖浓度范围：①1.7~2.8；②2.9~6.1③6.2~8.3④8.4~13.9⑤14.0~22.2 二、判断题（每题3分，共计18分） 1.凡是注日期的引用文件，其随后所有的修改单或修订版均适用于本标准。（F ） 2.准确度指的是一个测量值与可接受的参考值之间的相似程度。（F ） 3.系统准确度指的是一个测量系统得到的一组有代表性的测量结果与它们各自的参考值的一致性的接近程度。（T ） 4.精密度是指在规定条件下，相互独立的测试结果之间的一致程度。（T ） 5.重复性是在相同测量条件下，对同一被测量进行连续多次测量所得结果之间的一致性。（T ） 13.批（lot）为一个或多个器械组件或完整的器械，它们包含在不同的条件下生产的一个单一的类型、型号、类别、尺寸、组成或软件版本，并且在特定条件下，具有一致的特性和质量。（F ） 6.回收率是指在样品中加入定量标准品后，实际检测到的标准品加入量与已知标准品加入量之比，以百分数或小数表示。（T ）

危重患者血糖的监测及控制 习题及答案

危重患者血糖的监测及控制 姓名：科室：得分：一、选择题： 【A型题】 1、下列选项中不是高血糖发病机制的是：（ D ） A.胰岛素抵抗 B.应激 C.严重感染 D.肾上腺危像 E.胰岛B细胞功能丧失 2、胰岛素注射优先选择的部位是：（ A ） A.腹部 B.大腿内侧 C.臀部 D.手臂 E.静脉给药 3、以下哪种胰岛素属于速效胰岛素？（ B ）

A.诺和灵 R B.诺和锐 C.诺和灵N D.甘精胰岛素 E.甘舒霖N 4、下列哪项指标反映糖尿病患者近2-3月血糖控制不佳（ D ） A.尿糖阳性 B.血糖＞30mmol/L C.尿酮体强阳性 D.糖化血红蛋白A1为15.8% E.基础血浆C-肽水平低于400pmol/L 【B型题】 A.阴离子间隙增大 B.血浆乳酸5mmol/L C.尿糖阴性 D.空腹血糖9.1mmol/L E.血钠155mmol/L 5、高渗性昏迷（ E ）

6、糖尿病酮症酸中毒（ A ） 7、低血糖昏迷（ C ） 8、糖尿病（ D ） 9、乳酸性酸中毒（ B ） 【多选题】 10、下列关于重症患者血糖控制中正确的说法是：（ ABCDE ） A.解除应激因素 B.合理输注葡萄糖 C.静脉使用胰岛素 D.根据血糖监测情况调整胰岛素用量 E.一旦出现低血糖，静脉给予50%葡萄糖输入 二、简答题： 1、高血糖的危害： 答： 1.高血糖可以降低机体的免疫功能。2.高血糖也可降低补体的活性, 糖通过补体进行糖化作用, 和微生物竞争与补体的结合, 抑制调理作用。3.高血糖是加重机体能量代谢障碍的重要影响因素。高度应激状态下的高能耗、高氧耗和高分解代谢更加重了机体能量储备的耗竭, 最终可以导致细胞水肿、溶解和器官功能衰竭。2、患者发生高血糖的机制：

血糖监测技术进展

血糖监测技术进展 刘迎鸣 摘要：糖尿病作为一种流行的内分泌代谢系统疾病，血糖的监测对于糖尿病患者的诊断和治疗具有重要意义。本文对静脉血糖监测、毛细管全血血糖监测、无创血糖监测、动态血糖监测系统、人工胰岛系统的工作原理和优缺点以及研究现状进行了综述和分析，为临床选择血糖监测方法提供参考。 关键词：糖尿病血糖监测血糖仪 随着改革开放，中国经济取得了巨大的成功，人民生活水平快速提高。然而，人民生活方式改变和环境污染等因素的影响，糖尿病在中国已经由“富贵病”成为了一种慢性流行病。据2010年3月份新英格兰医学杂志发布的一份来自中国的研究报告，截至2008年5月份中国大陆已确诊的糖尿病人已达9240万，糖尿病前期病人达到1.46亿[1]。糖尿病已经成为发达国家继心血管病和肿瘤之后的第三大非传染性疾病，而在中国因糖尿病导致的并发症而死亡人数也日益增多。为了尽早的发现和控制糖尿病的发生、发展，必须开展血糖监测。 1、血糖监测的意义 糖尿病到目前为止还缺乏有效根治办法，患者一旦患病就会伴随终身，能否有效的控制糖尿病病情的发展是决定患者患病后生活质量以及护理成本关键因素。根据有关调查研究以及糖尿病控制与并发症试验结果显示，严格控制含糖物质的摄入量以及体内血糖水平，能够明显的降低糖尿病人的糖尿病并发症的发生和发展[2]。通过自我血糖监测（SMBG）可以客观的反应患者的血糖控制水平，评价用药效果，为医生有针对性的调整治疗方案提供了依据。同时，血糖监测可以作为糖尿病自我管理的一部分，可以帮助糖尿病患者更好的了解疾病状态，积极参与糖尿病的管理，及时调整用药，及时向医务工作者咨询的手段，从而提高治疗的依从性[3]。对高危人群进行筛查和定期血糖监测，可及时了解他们的情况，防病于未然。 2、血糖监测的主要方法 2.1静脉血浆葡萄糖生化测定 静脉血浆葡萄糖生化测定常采用葡萄糖氧化酶法或己糖激酶法，这些方法采用人体内静脉血血浆进行生化测定，测试结果比较准确、稳定、干扰性小，是临床检验中被广泛认可的常规手段。然而该方法的缺点在于速度慢，而且是有创检测、需血量大、有可能导致其他感

成人危重患者血糖监测和血糖控制评价的专家共识

成人危重患者血糖监测和血糖控制评价的专家共识 血糖监测与控制是ICU危重患者重要的诊疗内容。但当前血糖监测的方法很多，对血糖控制的评价缺乏统一的标准，造成对不同研究结果的解读困难。来自欧美国家的16位专家就成人危重患者血糖监测技术和血糖控制的评价达成共识。 1 血液取样 (1)对于病情危重需要实施有创血管内监测的患者：①所有血样应取自动脉导管；②如果暂时或持续不能获取动脉血，应留取静脉血样；③毛细血管血样(针刺法)不准确，应避免采用。 (2)对于病情相对较轻无需进行有创血管内监测的患者，可采用毛细血管血样(针刺法)。(3)临床研究的论文应报告取自动脉导管、中心和周围静脉导管或毛细血管(针刺法)血样的数量和比例。 2 ICU临床研究中血糖分析仪的选择 (1)取自动脉或中心静脉导管的血样应该在中心实验室或使用血气分析仪测定。血气分析仪是默认的血糖分析仪，中心实验室仅在报告不被耽搁的情况下使用。(2)对于未实施有创性血管内监测的患者，可以使用血糖仪测定毛细血管血样。(3)临床研究论文应报告使用中心实验室、血气分析仪或血糖仪测定血样的数量和比例。在所有情况下应报告所用测定仪的厂家、型号以及常规的校正和质量控制措施。 3血糖控制的报告——试验和观察性研究应报告内容 (1)集中趋势：对于一组患者血糖浓度的测量，应报告个体患者血糖平均值的中位数和四分位数值。(2)离散度：计算每位患者血糖浓度的标准差，然后报告该患者群体标准差的中位数和四分位数值。(3)低血糖：作为最低限度，研究者应报告至少经历1次中重度低血糖事件的患者的数量和比例[血糖浓度分别≤2．2 mmol／L(40 mg/dL)、2．3～3．9mmol/L(4l～

血糖检测仪 产品技术要求万孚

2.性能指标 2.1基本技术要求 2.1.1外观检查 应符合 GB/T 19634-2005 中 4.1 的要求。 2.2重复性 应符合 GB/T 19634-2005 中 4.2 的要求。 2.3准确度 对葡萄糖的回收率应为 80%-120%。 2.4质控物质 质控物质测量结果的95%应在血糖试条质控范围之内。 2.5功能 2.5.1仪器功能 2.5.1.1EC-101、EC-102、EC-103：可设置日期、时间、血糖单位、餐前/餐后预设、提醒血酮测试、高低血糖报警功能设置、定时闹钟提醒；自动存储最近测试的400个结果，并具备7天、14天、21天、28天检测结果平均值显示功能。 2.5.1.2EC-104、EC-105、EC-106：设置日期、时间、血糖单位、提醒血酮测试、高低血糖报警功能设置；自动存储最近测试的300个结果，并具备7天、14天、30 天、60天、90天检测结果平均值显示功能。 2.5.2提示功能 2.5.2.1EC-101、EC-102、EC-103：电量不足时进行提示；血糖测试结果低于1.1mmol/L（20mg/dL）或高于3 3.3mmol/L（600mg/dL）时提示“Lo”或“Hi”；餐前/餐后血糖标记提示；“#”符号特殊血糖标记提示；血酮测试提示；定时闹钟提醒；错误信息提示；血糖测试结果低于或高于用户设定的低/高警戒值时提示“低”或“高”。 2.5.2.2EC-104、EC-105、EC-106：电量不足时进行提示；血糖测试结果低于1.1mmol/L（20mg/dL）或高于3 3.3mmol/L（600mg/dL）时提示“Lo”或“Hi”；

重症患者血糖监测表格

重症患者血糖监测 1、血糖监测频率 对于存在初次静脉胰岛素；疑有低血糖；调整胰岛素用量；胃肠营养；床边血滤这些情况时，每0.5～1小时监测血糖一次。 （1）血糖≥ 8.3 mmol/L( 150mg/dl ）或 <3.9 mmol/L （70mg/dl）：每30分钟监测一次血糖 （2）滴注血管收缩药物时：每30分钟监测一次血糖 （3）血糖在3.9-8.3 mmol/L （70-150mg/dl）：每1小时监测一次血糖（4）血糖保持在3.9-6.1 mmol/L（70-110mg/dl）之间，并且胰岛素输注速度已4小时无需改变：每2小时监测一次血糖 2、起始的胰岛素用量 起始的胰岛素用量 血糖值（mmol/L）胰岛素用量 11.1～14.0 4 units/hour 14.1～17.0 6 units/hour 17.1～22.0 8 units/hour 3、胰岛素调整 胰岛素调整 血糖<4.0mmol/L停胰岛素泵。 血糖4.0-6.9mmol/L 下调1.0 U/h。 血糖较前上升1 mmol/L，胰岛素用量在原基础上上调0.5 U/h。 血糖较前下降1 mmol/L，胰岛素用量在原基础上下调0.5 U/h。

静脉注射胰岛素剂量与速率 ?患者的血糖水平比前一次降低大于1.2mmol/L小于5.5mmol/L时，不改变速度?血糖下降速度小于1.2mmol/L，或比以前更高时，应当提高胰岛素的输注速度（0.5～4u/h）； ?血糖下降速度大于6.5mmol/L时，重新调整胰岛素泵入速度，用量是先前的一半， ?达到目标血糖范围的低限时，以0.5～1u/h维持 ?若此过程中有含葡萄糖液体运用，可按3～6g增加1u胰岛素来控制

血糖监测

（八）血糖监测一、技术规范

二、操作流程 血糖监测操作考核流程 护士甲沟通

三、操作考核评分标准 项目标准 分值 质量标准 评分等级 A B C D 准备8分2 4 2 1.按规定着装，洗手、戴口罩。 2.用物准备齐全，放置合理有序。用物：治疗盘、 血糖仪、一次性采血针、血糖试纸、75％酒精、 棉签、弯盘、笔、医嘱执行单检查血糖仪的工作 状态、物品质量及有效期。 3.环境：清洁、舒适、注意保暖。 2 4 2 1 3 1 2 查对解释 评估10分2 3 3 2 1.查对患者及医嘱执行单。 2.解释操作目的、注意事项及配合指导，取得合作。 3.评估末端循环情况、是否存在影响血糖因素、采 集部位皮肤情况、有无酒精过敏史等。 4.询问患者大小便，必要时清洁双手。 2 3 3 2 1 2 2 1 1 1 试纸确认10分10 取出血糖监测仪，按下开机键，确认血糖监测仪显 示代码和试纸瓶标签上的代码相符。 10 5 0 0 采血 27分 5 5 17 1.床边核对患者及医嘱执行单，询问是否可以开始。 2.准备一次性采血针，从试纸瓶中取出试纸，盖好 瓶盖。 3.选择充血良好的手指末端，酒精棉签消毒手指， 确认酒精干透后将采血针去帽，采血端放在手指 侧面按按钮采血，待血液自然流出。（口述如出血 较少，应从远端轻轻挤压，不可直接挤压穿刺点， 否者会导致测量结果不准确）。 5 5 17 4 3 15 3 1 12 滴血17分17 棉签轻拭去第一滴血，将第二滴血滴入试纸测试区 的中央，直至试纸背面的确认窗变蓝，不可反复 滴入血液。指导患者按压采血点1～2分钟。 17 15 12 0 插入试纸，读取结果 10分10 将变蓝后的试纸插入血糖仪，约30秒后即显示测量 结果。读数，告知患者。结果如有异常，重复检 验一次，并告知医师，遵医嘱处理。 10 8 5 0 整理解释 记录10分2 2 3 3 1.再次核对医嘱、患者。 2.协助取舒适卧位→整理床单位。 3.询问患者感觉→告知注意事项。 4.正确处理用物→洗手→记录。 2 2 3 3 1 1 2 2 1 1 整体印象 3分 3 3 2 1 0 超时扣分全过程5分钟，超过10秒钟扣0.1分。 提问 5分 5 5 4 3 0 总分100

危重患者血糖监测及管理

ICU,重症监护室，很多人都会觉得这是一个离死神很近的地方，因为只有危重的病患才会被送到这里。在这里，病患的一切都由医护人员负责。护理人员不仅要确保病人的治疗状态正常，还要负责喂饭、安抚病人情绪、为病人做清洗工作，伺候病人大小便、要帮病人剪指甲等等，事无巨细。而恰恰就是这样一个离死神很近的地方，每天都在上演生命的感动，这份感动哪怕只是病患艰难的抬抬脚，只是轻轻地拨动智力珠盘……就是这些看似简单的运动，却唤起了生命重生的希望。 其实不然，这就是ICU，和以往大家心里想的重症监护室病区不同的是，住在这里的病患并非二十四小时只能躺在病床上，他们在接受日常临床治疗的同时，还可以适当的运动。而这些活动，就是我们护理人员给他们特别制定了一对一“私教”，以促进病人早期活动，减少肌肉萎缩、改善肌力和身体功能，为后期康复打下基础。 坐位活动，练习坐是重症患者康复的重要起点，应该尽早开始。从抬高床头30度，逐步过渡到床上坐位及床边端坐位。由于上身直立体位使回心血量减少，同时射血助力减低，因此心脏负荷低于卧位。为了增加患者对坐位训练的热情，护理人员还特别准备了一些智力玩具提供病人消磨时间。 护士长潘华良说，通过以上不同阶段的早期功能锻炼，促进了病患肌肉神经的功能恢复，提高了生活自理能力，降低ICU谵妄，缩短ICU住院时间，降低压疮及深静脉血栓等发生率，这些运动方案也得到了患者家属的一致好评。 先进的医学治疗为ICU里的危重病人挽回了生命，暖心的人文关怀让她们重拾了回归生活的信心。这就是医疗的一种高度和温度的有机结合，这也是我们一直在努力追求的优质医疗服务。 医护心声|总有一种感动让人坚持，爱在ICU 有人说ICU（重症医学科）是一个很有成就感的科室，可以见证生命奇迹；有人 说ICU是一个又脏又累死气沉沉的科室，每天都上演着生离死别；也有人说ICU 是一个精英部队，医务人员的应急能力个个出色，而我认为ICU是个严肃而冷静、威严而庄重、充满爱心与亲情的科室。 初到便深深爱上这里

危重患者血糖的监测及控制 习题及答案

危重患者血糖的监测及控制姓名：科室：得分：一、选择题： 【A型题】 1、下列选项中不是高血糖发病机制的是：（ D ） A.胰岛素抵抗 B.应激 C.严重感染 D.肾上腺危像 E.胰岛B细胞功能丧失 2、胰岛素注射优先选择的部位是：（ A ） A.腹部 B.大腿内侧 C.臀部 D.手臂 E.静脉给药 3、以下哪种胰岛素属于速效胰岛素？（ B ） A.诺和灵R B.诺和锐 C.诺和灵N D.甘精胰岛素 E.甘舒霖N 4、下列哪项指标反映糖尿病患者近2-3月血糖控制不佳（D ） A.尿糖阳性 B.血糖＞30mmol/L C.尿酮体强阳性 D.糖化血红蛋白A1为15.8% E.基础血浆C-肽水平低于400pmol/L 【B型题】

A.阴离子间隙增大 B.血浆乳酸5mmol/L C.尿糖阴性 D.空腹血糖9.1mmol/L E.血钠155mmol/L 5、高渗性昏迷（E ） 6、糖尿病酮症酸中毒（A ） 7、低血糖昏迷（C ） 8、糖尿病（D ） 9、乳酸性酸中毒（B ） 【多选题】 10、下列关于重症患者血糖控制中正确的说法是：（ABCDE ） A.解除应激因素 B.合理输注葡萄糖 C.静脉使用胰岛素 D.根据血糖监测情况调整胰岛素用量 E.一旦出现低血糖，静脉给予50%葡萄糖输入 二、简答题： 1、高血糖的危害： 答：1.高血糖可以降低机体的免疫功能。2.高血糖也可降低补体的活性, 糖通过补体进行糖化作用, 和微生物竞争与补体的结合, 抑制调理作用。3.高血糖是加重机体能量代谢障碍的重要影响因素。高度应激状态下的高能耗、高氧耗和高分解代谢更加重了机体能量储备的耗竭, 最终可以导致细胞水肿、溶解和器官功能衰竭。 2、患者发生高血糖的机制： 答：高血糖是由于激素和细胞因子水平增加引起的。1)升糖激素在应激状态下, 下丘脑-垂体-肾上腺轴(HPA)过度兴奋,促分解激素, 如糖皮质激素、胰升糖素、生长激素、儿茶酚胺等分泌增多, 而胰岛素分泌相对减少, 胰升糖素/胰岛素比例失调, 从而使糖异生增加, 肝糖原和肌糖原分解增加, 糖的生成速率明显增加。2)细胞因子在严重应激状态下, 来自不同组织的多种细胞因子对高血糖的产生也具有十分重要的作用。已知细胞因TNF-α主要通过调节胰岛素受体后信号的传导, 造成肝脏和骨骼肌对胰岛素耐受而使血糖增高。

危重患者血糖的监测及控制习题及答案新编完整版

危重患者血糖的监测及 控制习题及答案新编 HEN system office room 【HEN16H-HENS2AHENS8Q8-HENH1688】

危重患者血糖的监测及控制姓名：科室：得分： 一、选择题： 【A型题】 1、下列选项中不是高血糖发病机制的是：（ D ） A.胰岛素抵抗 B.应激 C.严重感染 D.肾上腺危像 E.胰岛B细胞功能丧失 2、胰岛素注射优先选择的部位是：（ A ） A.腹部 B.大腿内侧 C.臀部 D.手臂 E.静脉给药 3、以下哪种胰岛素属于速效胰岛素（ B ） A.诺和灵 R B.诺和锐 C.诺和灵N D.甘精胰岛素 E.甘舒霖N 4、下列哪项指标反映糖尿病患者近2-3月血糖控制不佳（ D ） A.尿糖阳性 B.血糖＞30mmol/L C.尿酮体强阳性 D.糖化血红蛋白A1为% E.基础血浆C-肽水平低于400pmol/L 【B型题】

A.阴离子间隙增大 B.血浆乳酸5mmol/L C.尿糖阴性 D.空腹血糖L E.血钠155mmol/L 5、高渗性昏迷（ E ） 6、糖尿病酮症酸中毒（ A ） 7、低血糖昏迷（ C ） 8、糖尿病（ D ） 9、乳酸性酸中毒（ B ） 【多选题】 10、下列关于重症患者血糖控制中正确的说法是：（ ABCDE ） A.解除应激因素 B.合理输注葡萄糖 C.静脉使用胰岛素 D.根据血糖监测情况调整胰岛素用量 E.一旦出现低血糖，静脉给予50%葡萄糖输入 二、简答题： 1、高血糖的危害： 答： 1.高血糖可以降低机体的免疫功能。2.高血糖也可降低补体的活性, 糖通过补体进行糖化作用, 和微生物竞争与补体的结合, 抑制调理作用。3.高血糖是加重机体能量代谢障碍的重要影响因素。高度应激状态下的高能耗、高氧耗和高分解代谢更加重了机体能量储备的耗竭, 最终可以导致细胞水肿、溶解和器官功能衰竭。 2、患者发生高血糖的机制： 答：高血糖是由于激素和细胞因子水平增加引起的。1)升糖激素在应激状态下, 下丘脑-垂体-肾上腺轴(HPA)过度兴奋,促分解激素, 如糖皮质激素、胰升糖素、生长激素、儿茶酚胺等分泌增多, 而胰岛素分泌相对减少, 胰升糖素/胰岛素比例失调, 从而使糖异生增加, 肝糖原和肌糖原分解增加, 糖的生成速率明显增加。2)细胞因子在严重应激状态下, 来自不同组织的多种细胞因子对高血糖的产生也具有十分重要的作用。已知细胞因TNF-α主要通过调节胰岛素受体后信号的传导, 造成肝脏和骨骼肌对胰岛素耐受而使血糖增高。

血糖监测技术

第二节血糖监测技术 【评估和观察要点】 1.评估患者 （1）评估患者的病情、意识、合作能力及身体状况。 （2）向患者解释监测血糖的目的、操作过程及如何配合，取得患者合作。 （3）患者身体是否存在影响血糖的因素，如有无剧烈运动、抽烟和饮用刺激性饮料。 （4）患者服用降糖药情况。 （5）评估患者手指皮肤情况、血液循环情况，有无对酒精过敏。 （6）患者进食情况，确认患者是否符合空腹、餐后2小时、随机血糖测定的要求。 2.评估血糖仪 （1）评估血糖仪的工作状态 （2）检查质控品有效期，贮存是否恰当。 （3）检查试纸条的有效期，贮存是否恰当，确认血糖仪上的号码与试纸号码一致。 （4）清洁血糖仪。 【护理要点】 （一）操作前准备 1.环境准备病房安静、整洁、整齐，光线充足。 2.护士准备着装整齐，洗手，戴口罩。 3.用物准备治疗盘（内放75%酒精、棉签）、血糖仪（密码与试纸一致）、 试纸（在有效期内）、记录单、一次性采血针。 （二）操作要点 1.核对医嘱 2.携用物至病人床旁，查对，做好解释，取得合作。 3.协助病人准备并清洁好双手，取舒适体位。 4.采血手下垂摆动10次促进血液循环（冬天采用此法，夏天可略）。 5.用75%酒精消毒采血部位，待干。 6.插入试纸条——开机。 7.绷紧皮肤，采血针紧贴皮肤按下。 8.弃去第一滴血液(用棉签抹去)，用第二滴血充满试纸的指定区域。 9.用无菌干棉签按压穿刺处,直至不出血。 10.再次查对, 读数并告知患者。 11.推出试纸——关机。 12.整理床单位，交待注意事项。

13.按院感要求处理用物，清洁血糖仪，洗手。 14.记录:包括被测试者姓名、测定日期、时间、测定结果、单位、检测 者签名等。 【健康指导】 1.告知患者血糖监测目的，取得合作。 2.指导末梢循环差的患者将手下垂摆动。 3.对需要长期监测血糖的患者，指导患者掌握自我监测血糖的技术和注意 事项。 【注意事项】 1.严格无菌技术操作.测血糖前，确认血糖仪上的号码与试纸号码一致。 2.采血前局部加温或手臂下垂以增加采血量。 3.快速血糖测定的为末梢毛细血管全血的血糖值，部位通常采用指尖、足 跟两侧,水肿或感染的部位不宜采血。 4.针刺部位尽量不选择指腹,应在手指尖两侧，部位要交替更换。 5.根据病人皮肤情况选择针刺深浅度。 6.采血时禁止过分挤压,应从掌根向指尖挤，切忌挤压针刺处，以防挤出 组织液影响血糖结果。 7.不要触摸试纸的滴血区、测试区。滴血量。应使试纸测试区完全变成红 色。 8.确认患者手指酒精干透后实施采血。 9.严重贫血、水肿、脱水、末梢循环不良及采血部位损伤的均影响结果。 10.某些药物，如扑热息痛、多巴胺、维生素C、甘露醇等对快速血糖仪 的检测存在干扰。 11.定期清洗和校对血糖仪。 12.试纸应在阴凉、干燥、避光、密封下保存，避免污染。 13.血糖仪应按卫生部文件及生产商使用要求,定期进行标准液校正、室 间质评和结果比对。 14.出现血糖异常（过低或过高）结果时应重复检测一次，立即通知医 生采取相应措施，必要时复检静脉生化血糖。

血糖监测的意义

血糖监测的方法与意义 一、血糖监测的概念及重要性 血糖监测是糖尿病管理中重要组成部分，其结果有助于评估糖尿病患者糖代谢絮乱的程度，制定合理的降糖方案，同时反应降糖治疗的效果并指导降糖方案的调整。目前临床监测血糖的途径有：毛细血管血糖、静脉血糖和组织间液血糖监测。其监测方式包括：便携式血糖检测仪、动态血糖检测仪、糖化血清清蛋白、糖化血红蛋白的测定。便携式的血糖检测仪反映的是即刻的血糖水平，它与动态血糖检测仪还可以反映血糖的波动情况和监测低血糖的发生，是“点”；糖化血清清蛋白和糖化血红蛋白是判断糖尿病长期控制血糖总体水平的重要指标，是“线”。只有通过点与线的结合，才能了解某些特定血糖的监测情况，有了解其在某一时期的总体水平。 二、SMBGS频率及时间 1、中国2型糖尿病防治指南推荐：1）使用胰岛素治疗的患者，在治疗 的开始阶段每天患者至少自我血糖监测5次，达到监测目标后可每天监 测血糖2-4次。2）非胰岛素治疗的患者，在治疗开始阶段每周3天，5-7 次/天，达到治疗目标后可每周监测3天，2次/天。3)若患者的血糖控制 较差或病情危重时，则应每天监测4-7次，直到病情稳定、血糖得到控 制为止；患者的病情稳定或已达血糖控制目标时，则每周监测3天，2 次/天。 2、SMBGS监测时间：可选择一天中不同的时间点，包括餐前、餐后2 小时、睡前及夜间。 1）空腹血糖：可以反映头天晚上的用药是否可以控制血糖到次日晨(即降糖药的远期疗效)，还可以间接反映机体自身基础胰岛素的分泌情况。 这里说的空腹血糖是指禁食8～12小时后的血糖，即清晨空腹状态下的 血糖，午餐和晚餐前的血糖不在此列。对于长期使用降糖药的患者来说， 空腹血糖的良好控制具有重要意义，这是因为血糖受多种因素的影响， 清晨空腹检查时能较大程度地排除这些影响，反映真实病情。测空腹血 糖最好在清晨6:00~8:00取血，采血前不用降糖药、不吃早餐、不运动。 2）午餐、晚餐前血糖：可用来指导患者调整进食量和餐前注射胰岛素(或口服降糖药)的量。 3）三餐后血糖：可以反映饮食控制和用药后的综合治疗效果，便于指导饮食和药物治疗；还可以间接反映进餐刺激后胰岛素的分泌情况。 对多数2型糖尿病患者来说，餐后两小时血糖有时比空腹血糖更重要， 因为这类患者空腹血糖可能并不高，但因其胰岛素分泌功能已经受损， 受高糖刺激后反应较差，而出现餐后高血糖。 应注意，测定餐后两小时血糖应从吃第一口开始到满两小时为止，有些 人从吃完饭开始计时，其结果就有了明显的差别。 4）睡前血糖：反映胰岛β细胞对进食晚餐后高血糖的控制能力。监测睡前血糖主要是为了指导病人科学加餐！ 5）夜间血糖：胰岛素治疗已接近达标，但空腹血糖仍高者；或疑有夜间低血糖者。

重症血糖监测与控制

重症血糖监测与控制 血糖升高是重症患者常见的病理现象。多年来，诸多的基础和临床研究得到几乎一致的认识，就是高血糖是危重患者的独立死亡危险因素之一；而通过合理或严格血糖控制，可显著降低多种疾病并发症的发生率和病死率。 （一）重症高血糖的病因与发病机制 重症患者的高血糖大致分为以下几种情况： 1．应激性高血糖：创伤、感染、手术、休克等应激等应激状态下，均可诱发血糖升高的病理现象称为应激性高血糖（SHG）。 2．原发或继发内分泌性疾病：与垂体有关的肢端肥大症，肾上腺相关的柯兴氏综合症，胰腺疾病，肿瘤移位内分泌等，均可影响胰岛素分泌、代谢和拮抗，造成血糖升高。 3．医源性高血糖：治疗中的含糖液输入过多，或器官功能障碍不能代谢造成血糖升高。亦有许多影响糖代谢并促使血糖升高的治疗或抢救用药，包括：皮质激素、生长激素、血管活性药物，儿茶酚胺，及噻嗪类利尿剂等。 （二）应激性高血糖危险因素 ①糖尿病，包括隐性糖尿病，有家族性糖尿病者应警惕，可能存在胰岛素相对或绝对不足，胰岛素抵抗和肝糖元异生增加的病理基础；②使用引起血糖升高的药物，如外援性儿茶酚胺、肾上腺和去甲肾上腺素；③胰岛素相对缺乏者，如老年患者、肥胖、急性胰腺炎、严重感染、低温、尿毒症、肝硬化和低氧血症等；④糖摄取不完全，主要为糖摄入过多，输入大量的碳水化合物或长期卧床糖利用相对过低所致。 （三）血糖监测 应激引起的血糖升高常与损伤的严重程度相关，是判断预后的重要指标。危重病期间的血糖升高随着病情的波动、加上治疗等多种因素干扰，常使血糖出现较大起伏，增加了治疗与控制难度，需短时间内的反复快速血糖监测，便于了解代谢和治疗进展，并指导于治疗调整，血糖监则成为非常重要的环节。实施胰岛素治疗方案中必须具备严格的血糖监测措施，避免发生低血糖反应和随后的反应性高血糖。特别对镇静状态或缺少对低血糖反应的神智不清患者、夜间无症状的低血糖反应和高血糖患者，更应重视血糖监测必要性。 血糖值的换算 血糖值表示法有两种单位,一种是毫克/分升（mg/dl），为旧制单位；另一种为毫摩尔/升（mmol/L）,为新制单位。现虽提倡用新制单位，但旧制单位仍在一定范围使用。所以,知道二者之间如何转换就很必要了。两种单位的换算公式为：mg/dl÷18=mmol/L； mmol/L×18=mg/dl。比如120mg/dl换算成以mmol/L为单位的数值时，需除以18,即 120mg/dl÷18=6.67mmol/L；6.67mmol/L换算成以mg/dl为单位的数值时，需乘以18,即6.67mmol/L×18=120mg/dl。 （四）重症血糖控制 重症患者的血糖控制的关键是胰岛素的应用。强化胰岛素治疗可有效的降低危重患者多种并发症的发生率和死亡率。 1．重症医疗单位工作前，应当认真对所有参与医护人员进行严格培训，并制定适合本医疗病员群体的血糖控制方案，提高对低血糖认识与防治策略。2005年我国重症医学营养指南将目标血糖设定为小于150mg/dl（8.3mmol/L）。

近红外光谱无创血糖检测技术研究全解

近红外光谱无创血糖检测技术研究 姓名:雷鹏 学号:2013022057 专业:光学

摘要:红外技术（Infrared Technique）是指以红外线的物理特性为基础。红外线是由于物质内部带电微粒的能全发生变化而产生的，它是一种电磁波．处于可见光谱红光之外．突出特点是热作用显著。红外线的波长介于可见光与无线电波之间．从0 .75μm～1000μm，可分为四个波段：近红外（0.75～3μm）、中红外（3～6μm）、远红外（6～15μm）、和极远红外（15～1000μm），红外线具光电效应，红外辐射效应，红外反射效应，大气传输特性等，这些特性为红外技术的应用 CH）振动的合频和各级倍频的吸收区一致，通过扫描样品的近红外光谱，可以得到样品中有机分子含氢基团的特征信息，而且利用近红外光谱技术分析样品具有方便、快速、高效、准确和成本较低，不破坏样品，不消耗化学试剂，不污染环境等优点，因此该技术受到越来越多人的青睐。糖尿病是一组以高血糖为特征的代谢性疾病，其检查主要靠血糖含量判断，本文提出一种利用近红外光谱进行无创血糖含量的方法，从而能够对糖尿病进行判断。 关键词：红外技术近红外光谱血糖

一、前言 红外技术的英文名称是：Infrared Technique。红外技术的内容包含四个主要部分： 1.红外辐射的性质，其中有受热物体所发射的辐射在光谱、强度和方向的分布；辐射在媒质中的传播特性--反射、折射、衍射和散射；热电效应和光电效应等。 2.红外元件、部件的研制，包括辐射源、微型制冷器、红外窗口材料和滤光电等。 3.把各种红外元、部件构成系统的光学、电子学和精密机械。 4.红外技术在军事上和国民经济中的应用。由此可见，红外技术的研究涉及的范围相当广泛，既有目标的红外辐射特性，背景特性，又有红外元、部件及系统；既有材料问题，又有应用问题 红外技术发展的先导是红外探测器的发展。1800年，F·W·赫歇尔发现红外辐射时使用的是水银温度计，这是最原始的热敏型红外探测器。 1830年以后，相继研制出温差电偶的热敏探测器、测辐射热计等。在1940年以前，研制成的红外探测器主要是热敏型探器。 19世纪，科学家们使用热敏型红外探测器，认识了红外辐射的特性及其规律，证明了红外线与可见光具有相同的物理性质，遵守相同的规律。它们都是电磁波之一，具有波动性，其传播速度都是光速、波长是它们的特征参数并可以测量。 20世纪初开始，测量了大量的有机物质和无机物质的吸收、发射和反射光谱，证明了红外技术在物质分析中的价值。 30年代，首次出现红外光谱代，以后，它发展成在物质分析中不可缺少的仪器。 40年代初，光电型红外探测器问世，以硫化铅红外探测器为代表的这类探测器，其性能优良、结构牢靠。 50年