S Panthakalam, D Bhatnagar, P Klimiuk
Department of Medicine, The Royal Oldham Hospital , Rochdale Road, Oldham, Lancashire
Correspondence to: E mail: d.bhatnagar@man.ac.uk
SMJ 2004 49(4): 139-141
Background and Aims: Corticosteroids are commonly used in the treatment of RA. Hyperglycaemia resulting from corticosteroid use can lead to problems, particularly in those with impaired glucose tolerance and diabetes mellitus. We have examined how rheumatology clinics monitored and managed hyperglycaemia at base line and during treatment of patients with rheumatoid arthritis on corticosteroids. Methods: Case notes of 102 patients with established RA, on long term steroids were reviewed. We recorded a) blood glucose level at base line and whilst on steroids b) when and whether hyperglycaemia was addressed. Results: There were 24 males and 78 females, with mean age of 62±15 years. Patients were on corticosteroids for a median duration of 24 months. Seventy-five per cent of patients were treated with oral prednisolone, the rest, except one patient on deflazacort, were on methyl prednisolone. Blood glucose was measured at baseline in 97% of patients with 37% and 38% being monitored at three months and six months respectively and 36% annually thereafter. Nine patients (8.8%) developed diabetes mellitus during treatment, but one patient was detected and managed. There were six patients with existing diabetes mellitus in whom glycaemic control worsened between three to six months, but only one patient had treatment adjusted. Conclusions and recommendations: Physicians need to be aware that corticosteroids can increase blood glucose, worsen preexisting diabetes and predispose to diabetes mellitus. Patients on long-term corticosteroids should be monitored at regular intervals as corticosteroid induced glycaemic excursions may lead to the development of diabetes mellitus and increased coronary risk.
Keywords: Rheumatoid arthritis (RA), hyperglycaemia, corticosteroid induced diabetes mellitus.
Corticosteroids are used in many rheumatological conditions, and they are particularly effective in rheumatoid arthritis.1 They can be delivered by a variety of means including oral, parenteral and intra-articular routes. In rheumatoid arthritis intra-articular steroids can help to settle a flare and bolus intravenous and intramuscularly steroids can be used as an adjunct to the slow acting second line drugs. Daily oral corticosteroid therapy has been shown to produce beneficial reduction in the radiological progression in rheumatoid arthritis.2 Significant corticosteroid induced hyperglycaemia is frequently encountered in patients with diabetes mellitus, but it is very likely also to occur in patients who have been normoglycaemic previously.3 There is no single mechanism by which glucocorticoids raise blood glucose levels. They do not appear to interfere with the cellular glucose uptake, but probably interfere with one or more intracellular mechanisms of glucose metabolism or insulin resistance.4 Insulin concentrations are increased with glucocorticoid administration especially in subjects who are previously normoglycaemic,5 but the increase is generally not sufficient to maintain normoglycaemia.
In patients without previous abnormalities of glucose tolerance, fasting blood glucose concentrations increase slightly; post prandial glucose increasing by a somewhat greater amount. These changes are accompanied by increased insulin production.6,7 In those with impaired glucose tolerance or diabetes mellitus, administration of glucocorticoids generally results in an increase in blood glucose concentrations.8 Patients with longer duration of diabetes mellitus may have greater incremental increase in hyperglycaemia as well.
The World Health Organisation has recently announced revised guidelines for the diagnosis of diabetes mellitus.9 These have been in practice for several years, but there are few data on whether rheumatologists are aware of the new criteria, by which diabetes mellitus is diagnosed at fasting serum glucose levels greater than 7mmol/l or a random greater than 11.1mmol/l. Such blood glucose values are not uncommon in patients attending rheumatology clinics who are on corticosteroids. In most patients glycaemic excursions are likely to be transient, but in some cases patients may be troubled with polyuria, polydipsia and repeated infections, in the short term. In the elderly there is always the risk of precipitating hyperglycaemic hyperosmolar states, including coma. In the long term, the overall burden of repeated increases in blood glucose may increase cardiovascular risk10 and the risk of micro vascular complications.11
We, therefore, examined whether patients with rheumatoid arthritis on long-term corticosteroid treatment, had their blood glucose estimated at base line and/or during treatment, and whether the hyperglycaemia was recognised and managed appropriately.
We examined the case notes of 250 patients with established rheumatoid arthritis attending The Royal Oldham Hospital rheumatology clinics between 1995 and 2000 to identify patients on long-term corticosteroids therapy. For each patient we documented the duration of rheumatoid arthritis, the presence of co-morbid conditions and the type, dose, duration and the method of administration of corticosteroid treatment. In addition we checked whether patients had blood glucose measurements carried out before or during corticosteroid therapy. Any incidental blood glucose measurements carried out whilst on corticosteroids (by general practitioners, and/or other physicians) were obtained from the Clinical Biochemistry laboratory computer records.
Patients
We were able to identify 102 patients with rheumatoid arthritis who were on corticosteroids. They included 24 males and 78 females with a mean age 62±15 (range: 37 to 92 years) years. The median duration of rheumatoid arthritis was 24 months (range 6 months to 120 months). Some patients had extra articular features such as Sjogren’s syndrome (n=2), pulmonary fibrosis (n=2), vasculitis (n=1) and episcleritis (n=1). Six patients were already known to have diabetes mellitus; four had type 2 diabetes mellitus treated with oral therapy, and two had type1 diabetes mellitus. All patients with diabetes mellitus (n=6) had good control as judged by home glucose monitoring and HbA1c (<7%). There were 15 smokers and five patients had co-existent hypertension, two patients had ischaemic heart disease and nine patients had bronchial asthma.
Duration and type of corticosteroid treatment
Patients were on corticosteroids for a median duration 24 months (range 6 to 60 months). Seventy-five per cent of patients were treated with oral prednisolone, the rest, except one patient on deflazocort, were on methyl prednisolone (table 1).
All patients were also on disease modifying antirheumatic drugs (DMARD) either as single or combination therapy (methotrexate n=53; sulphasalazine n=25; gold injections n=4; cholroquine n=9; leflunamide n=2; penicillamine n=2; cyclosporin A n=1; etanercept n=1)
Monitoring of blood glucose
Majority of patients (97.9%) had baseline laboratory estimation of blood glucose before initiating corticosteroid therapy. However, blood glucose had been monitored in 36 out of 102 (36.7%) at three months and 37 out of 102 (37.7%) at six months. At 12 months 62.2% (61/102) were still being monitored. Thereafter, 35.7% (35/102) had annual monitoring. All samples were taken as ‘random’ with no specimens taken when fasting or timed to meals.
Development of diabetes mellitus
Whilst on therapy, nine out of 102 patients (8.8%) developed diabetes mellitus (WHO criteria where with random blood glucose >11.1mmol/l or fasting blood glucose >7mmol/l), but only one patient was detected and managed. Four patients developed diabetes mellitus at six months and one at 12 months. Three patients developed diabetes mellitus after two years of corticosteroid therapy. Among the newly diagnosed nine patients with, six were on prednisolone 7.5mg/day, one each on five and 10 mg/day prednisolone respectively, and one patient received bolus methyl prednisolone on two occasions six months apart, but had not been on steroids previously. This patient developed diabetes mellitus a year later. In the absence of fasting blood glucose values and glucose tolerance tests it is difficult to comment on the extent of impaired glucose tolerance.
Management of patients with existing diabetes mellitus
All the six patients with existing diabetes mellitus had worsening of glycaemic control three to six months after corticosteroid therapy. HbA1c rose from 6.1±0.8% to 8.3±2.1% at six months (p=0.0427). Only in one patient, who was on insulin had treatment been adjusted to achieve normoglycaemia. In the rest it was not clear from the notes if treatment was adjusted.
In this retrospective study we found nearly 41% (n=102) of a cohort of 250 patients with rheumatoid arthritis were on treatment with corticosteroids for more than five years. The prevalence of diabetes mellitus was 8.8% compared to the expected population incidence of two to four per cent in the UK. In the absence of glucose tolerance tests and availability of fasting specimens it is difficult to know the prevalence of impaired glucose tolerance. It may be argued that the patients in our study could have developed diabetes mellitus even without the corticosteroid therapy due to unidentified risk factors. None of these patients had a family history of diabetes mellitus. In the clinical records there was no assessment of risk factors for diabetes mellitus before initiating corticosteroid therapy. Most clinicians are generally aware of the diabetogenic potential of corticosteroids, but our study highlights the lack of awareness of hyperglycaemia related problems with corticosteroids.
There are surprisingly few papers that actually document the prevalence of corticosteroid-induced hyperglycaemia. Much of the literature refers to steroid-induced diabetes mellitus and glucose intolerance in patients with renal transplants;12-18 however, the picture in post transplant diabetes is complicated by concomitant use of cyclosporin. Conn & Poynard have20 carried out a meta analysis of adverse events during steroid treatment, although the thrust of their analysis was on the prevalence of peptic ulceration. They report the odds ratio for developing diabetes mellitus as 1.7 (1.12 – 2.16 – 95% CI, p=0.02). However, the review does not clarify the definition of diabetes mellitus, and the presence of hyperglycaemia was ascertained in only four studies out of 93 that they reviewed. In a series 50 patients with asthma on oral corticosteroids for at least two years, the prevalence of glucose intolerance was reported as 28%.19 The large dose of steroids used in the above studies is a common theme. Patients with chest diseases and those who are on a post transplant immunosuppressive regime can be on doses as high as 30mg of prednisolone a day. This is in contrast to the use of steroids in rheumatoid arthritis where it is usual to administer doses around 7.5mg of prednisolone a day. In a case controlled review of pharmacy prescriptions in US21 there was some evidence of a dose dependent increase in the risk of developing diabetes mellitus. However, the variable measured was the risk of initiation of treatment for diabetes mellitus.
The paper perhaps underestimated the prevalence of corticosteroid-induced hyperglycaemia, because initiation of oral hypoglycaemic treatment would be dependent on physician awareness and monitoring of glycaemia during corticosteroid therapy. Our data are unable to provide information on the critical threshold of steroid use at which diabetes mellitus may develop. Ideally patients starting corticosteroids should have fasting glucose estimation or in those with a family history of diabetes mellitus a glucose tolerance test, prior to treatment. Once patients are started on steroids it would be worthwhile to carry out periodic home glucose monitoring, with patients instructed to contact the physician or nurse if certain cut off values are exceeded. Postprandial excursions in glucose are common following corticosteroids, and there may be a case for documenting pre-treatment timed two-hour postprandial glucose estimations.
There is a range of treatment options that can be used to control hyperglycaemia related to corticosteroids. Nonpharmacologic measures including consideration of reduction in steroid dose, a carbohydrate controlled diet, and exercise can be advised, but often the underlying condition requiring the use of corticosteroids generally makes this an impractical suggestion.
Rational pharmacologic therapy would ideally consist of prandial glucose regulators such as repaglanide and nateglinide, which are said to decrease postprandial glucose levels. However, given the degree of insulin resistance associated with corticosteroid use it may be more appropriate to use metformin or thizolidinediones such as rosiglitazone and pioglitazone.22 This may be particularly useful in those who may require long-term intermittent corticosteroid therapy. If the hyperglycaemia is severe then analogue insulins may be the insulin of choice, as they tend to produce greater reduction in postprandial glucoses than conventional non-analogue insulins.23
Our study suggests that clinicians involved in the care of patients with rheumatoid arthritis should be more aware of potential problems related to worsening glycaemia. Neither the diabetes nor rheumatology professional bodies have any guidelines to deal with corticosteroidinduced hyperglycaemia. Nevertheless, physicians and nurses in diabetes practice often have to deal with worsening hyperglycaemia due to corticosteroids use. There are no firm data documenting the extent of the problem, which must be common in specialities other than rheumatology too, such as chest medicine, haematology and obstetrics where corticosteroid use is not infrequent. Our findings indicate the need for a long-term prospective study to quantify the risk of development of hyperglycaemia in patients on long-term corticosteroid therapy, and to evaluate the hypoglycaemic agents that would be best suited for the management of corticosteroid-induced hyperglycaemia.
REFERENCES
1 Hoogwerf B, Danese RD. Drug selection and the management of corticosteroidrelated diabetes mellitus. Rheum Dis Clin North Am. 1999; 25: 489-505.
2 Van Riel PL, Haagsma CJ, Furst DE. Pharmacotherapeutic combination strategies with disease-modifying antirheumatic drugs in established rheumatoid arthritis. Baillieres Best Pract Res Clin Rheumatol. 1999; 13:689- 700.
3 Hirsch IB, Paauw DS. Diabetes management in special situations. Endocrinol Metab Clin North Am. 1997; 26: 631-45.
4 Pagano G, Cavallo-Perin P, Cassader M, Bruno A, Ozzello A, Masciola P, Dall’omo AM, Imbimbo B. An in vivo and in vitro study of the mechanism of prednisone-induced insulin resistance in healthy subjects. J Clin Invest. 1983; 72: 1814-20.
5 Kahn SE, Horber FF, Prigeon RL, Haymond MW, Porte D Jr. Effect of glucocorticoid and growth hormone treatment on proinsulin levels in humans. Diabetes. 1993; 42:1082-5.
6 Matsumoto K, Yamasaki H, Akazawa S, Sakamaki H, Ishibashi M, Abiru N, Uotani S, Matsuo H, Yamaguchi Y, Tokuyama K, Nagataki S. High-dose but not low-dose dexamethasone impairs glucose tolerance by inducing compensatory failure of pancreatic beta-cells in normal men. J Clin Endocrinol Metab. 1996; 81:2621-6.
7 Owen OE, Cahill GF Jr. Metabolic effects of exogenous glucocorticoids in fasted man. J Clin Invest. 1973; 52:2596-605.
8 Bruno A, Cavallo-Perin P, Cassader M, Pagano G. Deflazacort vs prednisone. Effect on blood glucose control in insulin-treated diabetics. Arch Intern Med. 1987; 147: 679-80.
9 Report of the Expert Committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 1998; 21(S): 5-19
10 The DECODE Study Group. Glucose tolerance and mortality: comparison of WHO and American Diabetes Association diagnostic criteria. The DECODE study group. European Diabetes Epidemiology Group. Diabetes Epidemiology: Collaborative analysis of diagnostic criteria in Europe. Lancet.1999; 354:617- 621
11 DCCT Research Group: The effect of intensive treatment on the development and progression of long-term complications in insulin dependant diabetes mellitus. N Engl J Med 1993; 329:977-986
12 Venstra D L et al. Incidence and cost of steroid side effects after renal transplantation. Transplantation Proceedings, 1999; 31:301-02
13 Fisher JE, Smith RS, Lagrandeur R, Lorenz RP.Gestational diabetes mellitus in women receiving beta-adrenergics and corticosteroids for threatened preterm delivery. Obstet Gynecol. 1997; 90: 880-3.
14 Silva F, Queiros J, Vargas G, Henriques A, Sarmento A, Guimaraes S. Risk factors for posttransplant diabetes mellitus and impact of this complication after renal transplantation. Transplant Proc. 2000; 32: 2609-10.
15 Hjelmesaeth J, Hartmann A, Kofstad J, Stenstrom J, Leivestad T, Egeland T, Fauchald P. Glucose intolerance after renal transplantation depends upon prednisolone dose and recipient age. Transplantation. 1997; 64: 979-83.
16 Ruiz JO, Simmons RL, Callender CO, Kjellstrand CM, Buselmeier TJ, Najarian JS. Steroid diabetes in renal transplant recipients: pathogenetic factors and prognosis. Surgery. 1973; 73: 759-65.
17 Arner P, Gunnarsson R, Blomdahl S, Groth CG.Some characteristics of steroid diabetes: a study in renal-transplant recipients receiving high-dose corticosteroid therapy. Diabetes Care. 1983; 6: 23-5.
18 McEvoy CE, Niewoehner DE. Adverse effects of corticosteroid therapy for COPD. A critical review. Chest 1997 111: 732-743
19 Lieberman P, Patterson R, Kunske R. Complications of long-term steroid therapy for asthma. J Allergy Clin Immunol. 1972; 49: 329-36.
20 Conn HO, Poynard T. Corticosteroids and peptic ulcer: meta-analysis of adverse events during steroid therapy. J Intern Med. 1994; 236: 619-32.
21 Gurwitz JH, Bohn RL, Glynn RJ, Monane M, Mogun H, Avorn J. Glucocorticoids and the risk for initiation of hypoglycemic therapy. Arch Intern Med. 1994; 154: 97-101.
22 Fujibayashi K, Nagasaka S, Itabashi N, Kawakami A, Nakamura T, Kusaka I, Ishikawa S, Saito T. Troglitazone efficacy in a subject with glucocorticoidinduced diabetes. Diabetes Care. 1999;22:2088-9
23 Roach P, Yue L, Arora V. Improved Postprandial Glycemic Control During Treatment With Humalog Mix25, a Novel Protamine-Based Insulin Lispro Formulation. Diabetes Care 1999;22:1258-1261