Best Practice for Diabetic Patients on Hemodialysis 2012
2015; Wiley; Volume: 19; Issue: S1 Linguagem: Inglês
10.1111/1744-9987.12299
ISSN1744-9987
AutoresToshiyuki Nakao, Masaaki Inaba, Masanori Abe, Kazo Kaizu, Kenji Shima, Tetsuya Babazono, Tadashi Tomo, Hideki Hirakata, Tadao Akizawa,
Tópico(s)Diabetes, Cardiovascular Risks, and Lipoproteins
ResumoGlycemic control in diabetic patients affects the development and progression of microvascular complications 1-3. A study showed that achieving HbA1c levels <6.9% (National Glycohemoglobin Standardization Program [NGSP] reference value) and 2-h postprandial plasma glucose levels <180 mg/dL prevents the development and progression of diabetic microvascular complications 2. Another study showed that intensive glycemic control with tight regulation of lipid levels and blood pressure decreases the incidence of cardiovascular events and improves the prognosis of diabetic patients 4. Intensive glycemic control at the initial stage of diabetes was also shown to reduce the incidence of cardiovascular events and improve the prognosis of diabetic patients in the following 10 years 5, 6. Thus, intensive glycemic control in diabetic patients is considered to prevent the development of not only microvascular but also macrovascular complications and improve their prognosis. Considering recent reports 7, 8, the prevention of hypoglycemia is also important. HbA1c level is the most common indicator for glycemic control in diabetic patients and represents a time-weighted mean plasma glucose level during the preceding 1 to 3 months. The life span of red blood cells is approximately 120 days. The plasma glucose level in the preceding 30 days contributes 50% to the HbA1c level, and those in the preceding 30–60 and 60–120 days contribute 25% each to the HbA1c level 9. For hemodialysis patients, however, the life span of red blood cells is shorter (approximately 60 days), blood loss and hemorrhage may occur during hemodialysis, and the percentage of immature red blood cells is increased by the administration of ESAs as a treatment for renal anemia. Therefore, hemodialysis patients tend to show low HbA1c levels for their glycemic control, which may underestimate glycemic control. GA is produced when serum albumin is glycated. The half-life of albumin is approximately 17 days. The GA level represents glycemic control in the preceding 2 to 4 weeks. The plasma glucose level in the 17 days immediately preceding blood sampling contributes 50% to the GA level, and those in the preceding 18–34 days and days before contribute 25% each to the GA level 9. Several studies showed that GA is not affected by the life span of red blood cells or ESA administration and can be used as a better indicator of glycemic control than HbA1c in hemodialysis patients 10-14. A study of Japanese patients compared hemodialysis patients with diabetes (n = 538), hemodialysis patients without diabetes (n = 828), and diabetic patients with normal kidney function (n = 365). The results showed that the HbA1c and GA levels in the hemodialysis patients with diabetes were significantly and positively correlated with the mean casual plasma glucose levels in the prior 3 months. However, the HbA1c levels in these patients were ∼30% lower than those in the diabetic patients with normal kidney function. The correlations between the mean casual plasma glucose levels and GA levels were similar in both groups of patients 12. Similar results were also found in US reports. A comparative study of diabetic patients without nephropathy (non-dialysis group, n = 49) and on hemodialysis (dialysis group, n = 258) showed significantly positive correlations between HbA1c and GA levels in both groups but showed a significant difference in the slope of the regression line for the two groups. The slope of the regression line showing the correlation between HbA1c and casual plasma glucose levels also significantly differed between the two groups. In that study, it was concluded that the HbA1c level for a given casual plasma glucose level is lower in the dialysis group than in the non-dialysis group. In contrast, there was no significant difference in the slope of the regression line showing the correlation between GA and casual plasma glucose levels in the two groups 13. Note that the GA level is dependent on factors other than plasma glucose level in the following groups: (i) patients with nephritic syndrome 15; (ii) patients undergoing peritoneal dialysis 16, 17; (iii) patients with thyroid dysfunction 18; and (iv) patients with hepatic cirrhosis 19. The majority of dialysis patients with diabetes have already developed microvascular and macrovascular complications, because they start dialysis as therapy for end-stage renal disease caused by diabetic nephropathy and they generally had diabetes for a long period by the time they start dialysis. Intensive glycemic control in such patients is considered to be effective for slowing the progression of not only microvascular complications such as retinopathy and neuropathy but also macrovascular complications and infectious diseases, and for improving the prognosis of these patients 20, 21. The prevention of hypoglycemia is also indispensable in dialysis patients with diabetes. Previous observational studies on the relationship between glycemic control and survival in hemodialysis patients with diabetes showed different results. In some studies, it was concluded that HbA1c level used as the indicator for glycemic control is predictive of survival in such patients 22-28. Others showed no such correlations in these patients 29-36, but their HbA1c levels were generally low. As mentioned above, the HbA1c level in hemodialysis patients underestimates glycemic control and cannot always be used as an accurate indicator of glycemic control. Therefore, the effects of glycemic control on the prognosis of hemodialysis patients should be reexamined using indicators other than HbA1c levels, such as plasma glucose and GA levels. An 11-year follow-up study of 245 new hemodialysis patients with diabetes in a Japanese facility showed that having (2-h postprandial) predialysis plasma glucose levels ≥180 mg/dL (the mean over the total observation period to death or termination of analysis) was associated with significantly shortened patient survival. However, there was no threshold HbA1c level at which survival shortened 20. Recently, a 6-year follow-up observational study of 54 757 hemodialysis patients with diabetes in the USA has shown higher mortality in the group with casual plasma glucose levels ≥200 mg/dL (patients categorized by plasma glucose level at increments of 25 mg/dL using mean casual plasma glucose levels of 150–175 mg/dL as the reference) 21. Moreover, survival was also shorter in the groups with HbA1c levels ≥8.0% and ≤6.9% (patients categorized by HbA1c level at increments of 1.0% using mean HbA1c levels of 7.0–7.9% as the reference). The distribution of the hazard ratio is V-shaped, indicating that HbA1c level is not directly associated with survival. Based on the above-mentioned studies, (predialysis) casual plasma glucose levels of <180–200 mg/dL are recommended as tentative targets for glycemic control. Although the GA level is a helpful indicator for glycemic control in hemodialysis patients as mentioned above, there have been a very limited number of reports on the relationship between the GA level and the risk of cardiovascular events or the prognosis of hemodialysis patients. A study of Japanese patients showed that the risk of cardiovascular events was significantly higher in diabetic hemodialysis patients with GA levels ≥23% 35. Another study of Japanese patients also showed that the survival rate of chronic hemodialysis patients with GA levels ≥29% was significantly low 34. In a recent 4-year follow-up study of 170 hemodialysis patients with diabetes in a Japanese facility, the relationship between glycemic control and survival in these patients was examined using the GA level as the indicator for glycemic control at the start of observation. The results showed that survival was longer in the patients who had no history of cardiovascular events as of the start of observation than in the patients with such a history. In addition, a report on the analysis of the patients who had no history of cardiovascular events revealed that survival was significantly longer in the group with GA levels 24.5% 37. In this report, however, the threshold GA level at which survival shortens was not found in the diabetic dialysis patients with a history of cardiovascular events as of the start of observation. A US study also showed significant associations of GA level with survival and the number of hospitalizations in hemodialysis patients with diabetes. However, the HbA1c level was not associated with survival or the number of hospitalizations in that study 38. Based on the above-mentioned studies, GA levels <20.0% are suggested as tentative targets for glycemic control in patients without a history of cardiovascular events. For patients with a history of cardiovascular events, however, GA levels <24.0% are suggested, because the positive effect of achieving GA levels <20.0% on the prognosis is not expected to exceed the negative effect of the increased incidence of hypoglycemia. Further study is required to definitively determine target values. The HbA1c level has long been used as an indicator in clinical practice and in many studies of glycemic control in dialysis patients with diabetes. However, as mentioned above, HbA1c levels are apparently low in hemodialysis patients and may not accurately represent glycemic control. Therefore, it is difficult to determine target HbA1c levels for glycemic control on the basis of the above studies. We calculated the mean HbA1c level in 195 hemodialysis patients with diabetes and found it to be 6.6% (NGSP reference value) for a predialysis plasma glucose level of 180 mg/dL. This value should be used only as a reference. Hemodialysis patients are prone to hypoglycemia because of various factors, and they require careful assessment of plasma glucose levels and appropriate management to prevent hypoglycemia. In glycemic control, multiple indicators, including (predialysis) casual plasma glucose and GA levels, should be comprehensively evaluated to reduce the risk of hypoglycemia and improve the prognosis of patients. Many of the recommendations in overseas guidelines for the treatment of diabetes in hemodialysis patients are based on weak evidence 39. The Japanese Society for Dialysis Therapy (JSDT) guide provides suggestions on the indicators for glycemic control and their tentative targets based on previous reports. Further clinical trials are needed to strengthen the basis of these suggestions. Self-monitoring of blood glucose (SMBG) is useful for maintaining and improving glycemic control in all patients with type 1 diabetes and patients with type 2 diabetes who are treated with insulin 2, 1. Therefore, all patients who are treated with insulin, irrespective of diabetes type, are recommended to carry out SMBG. Clinicians should also monitor their pre- and postdialysis plasma glucose levels in each hemodialysis session. When patients treated with insulin are suspected to have developed various symptoms resulting from hypoglycemia or feel sick, plasma glucose level should be monitored to understand their condition. Some studies showed that the adjustment of insulin dose based on the results of SMBG is indispensable for providing intensive insulin therapy to non-dialysis patients with type 1 diabetes and contributes to improved glycemic control and the prevention of complications 2, 1. SMBG is also an essential component of intensive insulin therapy for non-dialysis patients with type 2 diabetes and has similar contributions 2. There has been insufficient evidence on the frequency of monitoring plasma glucose level in non-insulin-treated patients, including non-dialysis patients. However, hemodialysis patients generally visit their medical institution three times per week and have more opportunities to know their plasma glucose levels than non-dialysis patients. This advantage enables a more accurate evaluation of glycemic control in hemodialysis patients. For hemodialysis patients with stable glycemic control who are taking oral hypoglycemic agents, predialysis plasma glucose level should be monitored once a week. For hemodialysis patients who have changed their dose of a hypoglycemic agent or switched to or added another hypoglycemic agent, predialysis plasma glucose level should be monitored more frequently until stable glycemic control is achieved. For diabetic hemodialysis patients who have achieved good glycemic control without drug therapy, predialysis plasma glucose level should be monitored at least once a month. When hyperglycemia is detected, clinicians are advised to reevaluate glycemic control in their patients. The GA level represents the mean plasma glucose level in the preceding 2 to 4 weeks, which is relatively shorter than for the HbA1c level. The GA level does not greatly change in 2 weeks when diabetic patients have stable glycemic control. Therefore, once-a-month monitoring of the GA level is recommended for diabetic hemodialysis patients. Currently, continuous glucose monitoring (CGM) systems can be used to continuously monitor the plasma glucose level to understand its dynamic changes. CGM systems are applicable to the evaluation of postprandial changes in plasma glucose level, the levels at nighttime and midnight, and the presence of asymptomatic hypoglycemia. In particular, the use of CGM systems for diabetic hemodialysis patients treated with insulin is attracting attention because circadian changes in plasma glucose level may differ between dialysis and non-dialysis days 40, 41. There are few reports on the incidence rate of diabetes mellitus after initiating of dialysis therapy. The American Diabetes Association recommends that regular testing should begin at age ≥45 and should be considered in all adults who have obesity and who have one or more risk factors: physical inactivity, family history of diabetes, women who were diagnosed with gestational diabetes mellitus or who delivered a large baby, hypertension, dyslipidemia, women with polycystic ovary syndrome, other clinical conditions associated with insulin resistance, or history of cardiovascular disease or abnormal glucose tolerance 42. End-stage kidney disease results in increased insulin resistance, which has been reported as a predictor of cardiovascular events 43-45. Therefore, casual plasma glucose and GA levels should be monitored at least once a year in hemodialysis patients without diabetes as a screening for type 2 diabetes. When a patient is suspected to have diabetes, clinicians should make a diagnosis following the clinical guidelines established by the Japan Diabetes Society 46, which recommend that both plasma glucose and HbA1c levels should be monitored in 1-day blood sampling. Criteria for the diagnosis of diabetes are as follows: fasting plasma glucose ≥126 mg/dL; 2-h plasma glucose ≥200 mg/dL during an oral glucose tolerance test (OGTT) using a glucose load containing the equivalent of 75 g of glucose; or casual plasma glucose levels ≥200 mg/dL; and HbA1c levels ≥6.5% 46. When patients satisfy the criteria for both plasma glucose and HbA1c levels using blood sampled on 1 day, they are diagnosed as having diabetes by only one blood sampling. As mentioned above, the HbA1c level tends to be low in hemodialysis patients and underestimates glycemic control. Thus, some hemodialysis patients who should be diagnosed as having diabetes may have HbA1c levels <6.5% but satisfy the plasma glucose level criterion for diabetes. Therefore, testing (monitoring of fasting and casual plasma glucose levels and OGTT) should be repeated in the case of diabetes diagnosis of hemodialysis patients. As of October 2011, approximately 25 types of dialysis fluid were commercially available from four companies in Japan. Table 1 shows a summary of the composition of each dialysis fluid. The glucose level in the dialysis fluid is 0, 100, 125, or 150 mg/dL. Many dialysis patients with diabetes have already developed hyperglycemia before the initiation of dialysis. Generally, morning hemodialysis starts between 8:30 and 9:00, which is 1 or 2 h after breakfast and when the plasma glucose level reaches the maximum. When hemodialysis is started in patients with such a high plasma glucose level using a dialysis fluid with a glucose level in the range of 100–150 mg/dL, the large difference in glucose level between the blood and the dialysis fluid allows plasma glucose to diffuse into the dialysis fluid, decreasing the plasma glucose level 47, 48. When hemodialysis is performed using a dialysis fluid with a glucose level of 100 mg/dL, the intradialytic decrease in plasma glucose level is greater in insulin-treated patients with higher predialysis plasma glucose levels (Fig. 1) 48. For patients with good glycemic control (i.e. diabetic patients with low predialysis plasma glucose levels), the plasma glucose level tends to decrease but not significantly. The plasma glucose level in dialysis patients without diabetes rarely changes during dialysis. Changes in plasma glucose level are more significant when a dialysis fluid with a plasma glucose level of 0 mg/dL is used. Plasma glucose levels monitored at the inlet and outlet of a dialyzer during hemodialysis similarly decrease at the start of the hemodialysis session and 2 and 4 h after the start. The degree of decrease is large in patients with extremely high predialysis plasma glucose levels. These trends are similarly observed in patients with both good and poor glycemic control. This is because glucose in the blood diffuses into the dialysis fluid in accordance with the predialysis plasma glucose level during hemodialysis. Changes in plasma glucose level during hemodialysis session using glucose-containing or non-glucose-containing dialysis fluid 48. DM, diabetes mellitus; glucose(+), glucose-containing dialysate (100 mg/dL); glucose(–), non-glucose-containing dialysate (0 mg/dL); HD, hemodialysis; In, insulin; PG, plasma glucose; N.S., not significant; 2°, 2 h after start of hemodialysis session; 4°, 4 h after start of hemodialysis session. The plasma glucose level in diabetic patients changes even after hemodialysis, because of the effect of dialysis fluid 47, 48. In patients with markedly decreased plasma glucose levels during a hemodialysis session, their plasma glucose levels are greatly elevated after dialysis or lunch, causing significant hyperglycemia (Fig. 2). In hyperglycemic patients, their plasma glucose levels are not markedly elevated even after lunch on a non-dialysis day, because their plasma glucose levels do not decrease. In diabetic patients who have good glycemic control, their plasma glucose levels do not decrease during dialysis and hyperglycemia does not develop after hemodialysis. Hyperglycemia that develops after hemodialysis as a result of the decrease in plasma glucose level during hemodialysis is called hemodialysis-induced hyperglycemia. It is speculated that when the plasma glucose level decreases, insulin secretion decreases, but the secretion of counter-regulatory hormones that elevate plasma glucose levels increases to maintain a stable plasma glucose level. Glucose-level-elevating hormones, such as glucagon, may contribute to the development of hemodialysis-induced hyperglycemia. Circadian changes in plasma glucose level in dialysis patients with diabetes on dialysis and non-dialysis days 48. HD, hemodialysis; SD, standard deviation. *P < 0.05, **P < 0.01. Patients undergoing insulin therapy generally receive insulin before the hemodialysis session. Insulin probably contributes to the intradialytic decrease in plasma glucose level, but this contribution may weaken during hemodialysis, because insulin in the blood is removed by hemodialysis and the blood insulin level decreases to various degrees. However, insulin is still one of the factors affecting the intradialytic decrease in plasma glucose level. Hyperglycemia at the start of hemodialysis If marked hyperglycemia ≥500 mg/dL was observed at the start of each dialysis session, subcutaneous injection of a small dose (2–4 units) of ultrafast-acting insulin is suggested. In these cases, levels of plasma glucose should be again monitored in 2 h. The target plasma glucose levels are 100–249 mg/dL. Sudden and excessive decrease in plasma glucose after insulin injection (<100 mg/dL) should be avoided. When plasma glucose ≥600 mg/dL is detected, urgent measures of blood gas analysis, serum potassium, and if possible, ketone bodies in the blood are recommended, to exclude the diagnosis of diabetic ketoacidosis (DKA). Patients diagnosed as having DKA should be immediately hospitalized. When hyperglycemia occurs frequently, consultation with a diabetes specialist is strongly recommended to fundamentally revise the current diabetes therapeutic regimen. To the best of our knowledge, there are no reports on the management of hyperglycemic episodes detected at the start of a dialysis session. The plasma glucose level naturally decreases after the start of a hemodialysis session because of the diffusion of glucose into dialysis fluid. Data from a Japanese facility showed that in patients who had predialysis plasma glucose levels <500 mg/dL, their plasma glucose levels decreased to <300 mg/dL in a 3–4-h hemodialysis session without the use of hypoglycemic agents (Table 1). Therefore, hyperglycemic patients with predialysis plasma glucose levels <500 mg/dL do not usually need insulin injection. However, for significantly hyperglycemic patients with predialysis plasma glucose levels ≥500 mg/dL, subcutaneous injection of a small dose (two to four units) of ultrafast-acting insulin is suggested. In this case, plasma glucose level should be again monitored 2 h. The target plasma glucose levels are suggested to be 100–249 mg/dL. Changes in the plasma glucose level caused by insulin differ among patients. Even a small dose of insulin, as mentioned above, may markedly decrease the plasma glucose level. Therefore, care should be taken to prevent a sudden or extreme drop in the plasma glucose level (<100 mg/dL). When hyperglycemia occurs frequently, consultation with a diabetes specialist is strongly recommended to fundamentally review the current diabetes treatment regimen. Hemodialysis patients are considered to rarely develop DKA. However, several case report of Japanese patients (Table 2) 49-62 showed that DKA patients on hemodialysis had higher plasma glucose levels (all patients, ≥600 mg/dL; average, 1336 ± 369 mg/dL) than those without renal disease and developed an advanced metabolic acidosis and hyperkalemia, which cannot be explained by renal disease alone. When hyperglycemia with plasma glucose levels ≥600 mg/dL is detected, urgent measures, such as blood gas analysis, serum potassium monitoring, and if possible, monitoring of ketone bodies in the blood, are recommended to rule out DKA. Patients diagnosed as having DKA should be immediately hospitalized. Hypoglycemia before and after hemodialysis Immediate treatment is required in insulin-treated patients if predialysis plasma glucose levels are <60 mg/dL or hypoglycemic symptoms develop. If oral intake is possible, 5–10 g of glucose should be taken orally. Otherwise, 20 mL of 50% glucose solution should be injected via the hemodialysis circuit in around 60 s. Plasma glucose measurements should be repeated every 30 to 60 min thereafter. If plasma glucose levels decrease below 60 mg/dL, the above procedure should be repeated. If hypoglycemia was detected at the end of dialysis session, the procedure of administration of glucose with the monitoring of plasma glucose should be repeated to obtain stable normoglycemia before the end of the session. When hypoglycemia occurs frequently, consultation with a diabetes specialist is strongly recommended to fundamentally revise the current diabetes therapeutic regimen. Urgent treatment, as described above, are required for insulin-treated patients with predialysis plasma glucose levels <60 mg/dL or those with significant hypoglycemic symptoms. Hypoglycemic symptoms are classified into early-onset autonomic symptoms and late-onset central neurological symptoms. Many hemodialysis patients with diabetes have autonomic neuropathy and decreased activities of counterregulatory hormones such as glucagon 63. In these patients, autonomic symptoms, such as cold sweat, palpitations, and finger tremor, may manifest as the initial symptoms of hypoglycemia, rapidly leading to neuroglycopenia symptoms, such as abnormal behavior, seizure, a reduced level of consciousness, and finally coma (asymptomatic hypoglycemia). Therefore, care should be taken for hemodialysis patients with diabetes. For patients with frequent hypoglycemic episodes, similarly to patients with frequent hyperglycemic episodes, consultation with a diabetes specialist is strongly recommended to fundamentally review the current diabetes treatment regimen. Sulfonylureas (SUs), biguanide, thiazolidinedione, nateglinide (a ultrafast-acting insulin secretagogue), and sitagliptin (a dipeptidyl peptidase-4 [DPP-4] inhibitor) are contraindicated in dialysis patients. The following oral hypoglycemic agents can be used in dialysis patients: mitiglinide and repaglinide (fast-acting insulin secretagogues); acarbose, voglibose, and miglitol (α-glucosidase inhibitors [α-GIs]); and vildagliptin, alogliptin, linagliptin, teneligliptin, and anagliptin (DPP-4 inhibitors). Dose of oral hypoglycemic agents Fast-acting insulin secretagogues: mitiglinide and repaglinide should be initiated at a low dose with caution. α-GIs: acarbose and voglibose can be used at up to a regular dose. Miglitol should be used with caution. DPP-4 inhibitors: vildagliptin should be initiated at a low dose (25 mg/day). Alogliptin (6.25 mg/day) and anagliptin (100 mg/day) should be used with caution. Linagliptin and teneligliptin can be used at a regular dose. Insulin is partly degraded in the kidney. Dialysis patients with impaired kidney function are prone to hypoglycemia, because the metabolism and excretion of insulin, as well as drugs, are delayed. Among oral hypoglycemic agents, SUs stimulate insulin secretion and tend to induce prolonged hypoglycemia (once developed, hypoglycemia is prolonged); hence, they are contraindicated in dialysis patients 64-67. If the sole use of a first-choice drug does not improve glycemic control, increasing the dose or adding another hypoglycemic agent with a different action mechanism should be considered. When glycemic control is still poor even with the use of multiple hypoglycemic agents, additional injection of basal insulin or switching to insulin therapy should be considered. It is necessary to avoid continuing the use of hypoglycemic agents if they fail to improve glycemic control. As shown in Table 1, there are six classes of oral hypoglycemic agents, which are categorized as (i) agents that stimulate insulin secretion (insulin secretagogues); (ii) agents that improve insulin resistance (insulin sensitizers); and (iii) agents that improve postprandial hyperglycemia (postprandial glycemic regulators) 68. Insulin secretagogues include SU, fast-acting insulin secretagogues, and DPP-4 inhibitors. Insulin sensitizers include biguanide and thiazolidine. Postprandial glycemic regulators include fast-acting insulin secretagogues and α-GIs 68. According to Japanese guidelines, SUs, biguanide, and thiazolidinediones are contraindicated in patients with severe renal impairment. α-GIs and DPP-4 inhibitors, as well as mitiglinide and repaglinide (which are fast-acting insulin secretagogues), can be used in dialysis patients. However, there is no clear evidence that determines the best agent among them. All SUs have a high protein-binding rate and are not removed by dialysis. Although their major metabolic pathway is the liver, SUs easily induce hypoglycemia in dialysis patients through the accumulation of active metabolites that decrease the plasma glucose level 69. Gliclazide is metabolized in the liver and 99% or more of its metabolites are excreted from the kidney (60–70%) and in the feces (10–20%). The use of gliclazide is recommended in the Kidney Disease Outcomes Quality Initiative guidelines because of the very low activity of its metabolites 70. However, it is stated as contraindicated in the explanatory leaflet of SUs in Japan 64. Fast-acting insulin secretagogues stimulate insulin secretion by a mechanism similar to that for SUs. However, they have a faster onset of action, a greater elevation of blood insulin level, and a shorter duration of lowering plasma glucose level (approximately 3 h) than SUs. As a side effect, they increase the risk of hypoglycemia. Although the risk of hypoglycemia posed by fast-acting insulin secretagogues is smaller than that posed by SUs 71, there is insufficient evidence that supports this. Nateglinide is contraindicated in dialysis patients, because its metabolites lower the plasma glucose level and are excreted from the kidney, which increases the risk of hypoglycemia in these patients 72, 73. Mitiglinide rarely increases the risk of hypoglycemia even in patients with renal impairment, because its metabolites have no plasma glucose-lowering effect 74. However, the half-life of mitiglinide in the blood is prolonged in patients with renal impairment, including dialysis patients. Therefore, mitiglinide should be initiated at a low dose (7.5–15 mg/day). Some studies showed that the plasma glucose and GA levels in Japanese hemodialysis patients were reduced by mitiglinide at a dose lower than the regular dose 75, 76. Repaglinide is excreted in the bile and its metabolites do not decrease plasma glucose level. Although repaglinide has been safely used in patients with renal impairment 77-79, it should be initiated at a low dose
Referência(s)