COPD, 00:1–6, 2014 ISSN: 1541-2555 print / 1541-2563 online Copyright © Informa Healthcare USA, Inc. DOI: 10.3109/15412555.2014.898052

ORIGINAL RESEARCH

Safety of Metformin in Patients with Chronic Obstructive Pulmonary Disease and Type 2 Diabetes Mellitus Andrew W. Hitchings,1 John R.H. Archer,1,2 Shelley A. Srivastava,1,3 and Emma H. Baker1

COPD Downloaded from informahealthcare.com by Kainan University on 04/08/15 For personal use only.

1

Clinical Pharmacology, Institute of Infection and Immunity, St George’s, University of London, London, UK

2

Clinical Toxicology, Guy’s and St Thomas’ NHS Foundation Trust, London, UK

3

Charing Cross Hospital, Imperial College Healthcare NHS Trust, London, UK

Abstract Type 2 diabetes mellitus (T2DM) is commonly associated with chronic obstructive pulmonary disease (COPD). Metformin is a valuable treatment for T2DM, and may offer additional benefits in COPD. However, due to its rare association with lactic acidosis, its safety in COPD is uncertain. We retrospectively identified patients with T2DM who had been admitted to hospital for COPD exacerbations. We compared those who were taking metformin with those who were not, with respect to their lactate concentration (primary endpoint) and survival (secondary endpoint). The study cohort (n = 130) had a mean (±standard deviation) age of 73.0 ± 9.8 years and 47 (36%) were female. Arterial blood gases were recorded in 120 cases: 88 (73%) were hypoxemic, 45 (38%) were in respiratory failure and 33 (28%) had respiratory acidosis. The 51 patients (39%) in the metformin group had a median (interquartile range) lactate concentration of 1.45 mmol/L (1.10–2.05) versus 1.10 mmol/L (0.80–1.50) in the non-metformin group (p = 0.012). Median survival was 5.2 years (95% CI 4.5–5.8) versus 1.9 years (1.1–2.6), respectively (hazard ratio 0.57; 95% CI 0.35–0.94). This remained significant in a multivariate model adjusted for measurable confounders. In conclusion, among patients with COPD at high risk for lactate accumulation, metformin therapy was associated with a minor elevation of lactate concentration of doubtful clinical significance. Metformin was associated with a survival benefit, but this must be interpreted cautiously due to possible effects from unmeasured confounders. Viewed collectively, the results suggest that COPD should not present a barrier to the investigational or clinical use of metformin.

Introduction

Keywords: adverse drug reaction, lactic acid, metformin, obstructive pulmonary diseases, Type 2 Diabetes Mellitus, survival Correspondence to: Dr. Andrew Hitchings, Division of Biomedical Sciences (mail point J1A), St George’s, University of London, Cranmer Terrace, London SW17 0RE, United Kingdom, phone: +44 (0)20 8725 5380, email: [email protected]

Type 2 diabetes mellitus is common among people with chronic obstructive pulmonary disease (COPD) (1). It is associated with significantly reduced lung function, exercise capacity and quality of life, and an increased risk of death (2, 3). Metformin is recommended as the first-line treatment for most patients with type 2 diabetes mellitus (4, 5), in whom it may reduce the risk of cardiovascular events and death (6, 7). In addition, a recent open-label study of metformin in patients with COPD has indicated that it may improve health status and respiratory muscle strength (8). A retrospective study has suggested it may improve forced vital capacity (9). Metformin also offers theoretical benefits in COPD due to its anti-inflammatory (10–12) and antioxidant (13, 14) effects, and in limiting glucose flux across airway epithelium in association with respiratory infection (15).

1

COPD Downloaded from informahealthcare.com by Kainan University on 04/08/15 For personal use only.

2

Hitchings et al.

Against this backdrop, an area of uncertainty remains. Metformin is rarely associated with lactic acidosis, which may be fatal (16, 17). Meta-analysis of clinical trial data suggests that, in general, the risk is extremely low (18). However, whether this finding can be extrapolated to patients with COPD is uncertain, because their representation in clinical trials of metformin has been limited and poorly reported (18). The British National Formulary (16) and a US Federal Drug Administration ‘black box’ warning (17) advise that metformin should be withheld promptly in the presence of any condition associated with hypoxemia. Appropriately or not, these factors restrict the clinical use of metformin among patients with diabetes and coexisting COPD, and may impede investigation of the potential benefits of metformin specifically in COPD. As a prelude to conducting a randomized, doubleblind, placebo-controlled trial, we sought to investigate the safety of metformin in COPD in a retrospective cohort, for which long-term follow-up data could be obtained.

Methods Study overview This was a retrospective observational study. Informed consent was not sought, as data were collected in parallel with an audit of prescribing practices. The National Information Governance Board confirmed that support under Section 251 of the NHS Act 2006 (to access identifiable data without consent) was not required. A favorable opinion was provided by the South West London Research Ethics Committee. Study population and data collection All admissions to St George’s Hospital, London between 1998 and 2010 with International Classification of Diseases (ICD-10) codes for both COPD (J41–J44) and type 2 diabetes mellitus (E10, E11, E14) were identified from clinical coding data. Patients were included if they were aged >40 years, had documented diagnoses of COPD exacerbation and type 2 diabetes mellitus, and a record of their drug treatment. Only first admissions for individual patients during the study period meeting these criteria were included. Data were extracted from the patients’ written and electronic case notes in relation to their co-morbidities, illness severity, relevant investigations, treatment and survival. Co-morbidity burden was quantified using the Charlson co-morbidity index (19) and acute illness severity by APACHE II score (Acute Physiology and Chronic Health Evaluation II (20)). Lactate concentration and indices of gas exchange and acid–base status were obtained from arterial blood analysis records. Glycemic control was assessed by hemoglobin (Hb) A1c levels, where a measurement had been made within 90 days of the index admission. Vital status and dates of death were ascertained from the hospital electronic record system, which is regularly synchronized with national

mortality data. Follow-up was deemed to be censored if the patient was alive at the time of the last electronic record review in September 2013.

Endpoints and group assignment The primary endpoint was the first plasma lactate concentration at hospital admission. The secondary endpoint was length of survival until death from any cause. To facilitate interpretation of results, we prospectively considered the minimum difference in plasma lactate concentration that would be likely to alter the management of a patient taking metformin. In the absence of empirical data addressing this point directly, we selected a threshold of 1.0 mmol/L on the basis of our experience. For comparisons of variables recorded at presentation, patients were divided into metformin and nonmetformin groups according to whether or not metformin was recorded in their admission drug history. For the analysis of deaths that occurred in hospital, group assignment was determined by whether metformin had been prescribed in the week preceding death. For deaths that occurred after hospital discharge, group assignment was based on the presence or absence of metformin in the discharge prescription. Statistical analysis Data are described by mean ± standard deviation; median (interquartile range); and proportions, as appropriate. Only those patients for whom the lactate concentration (the primary endpoint) had been measured were included in the primary endpoint analysis. All patients were included in the secondary (survival) analysis. Continuous data were compared using the t-test or Mann– Whitney test as appropriate; proportions were compared using Fisher’s exact test. Survival data were compared by Kaplan–Meier survival curves, log-rank tests, and Cox proportional hazards regression. All statistical tests were two-sided, with the significance level set at 0.05, and calculated using PASW Statistics 18 (SPSS Inc).

Results Patient characteristics Case notes were reviewed for 178 patients, of whom 130 met the inclusion criteria. The reasons for exclusion were: no documented physician diagnosis of COPD exacerbation (33 patients) or diabetes mellitus (8 patients); and incomplete drug history (7 patients). Fifty-one patients (39%) were on metformin and 79 (61%) were not. Patients on metformin were younger than those not on metformin (mean 70.5 ± 9.6 vs 74.7 ± 9.6 years, respectively; p = 0.016), and their body weight was higher (median 83.3 [interquartile range 71.4– 104.1] vs 67.3 [52.2–79.8] kg, respectively; p = 0.003). Arterial blood gas results were available for 120 patients (92%). Eighty-eight (73%) were hypoxemic and 33 (28%) had respiratory acidosis. Other relevant characteristics are detailed in Table 1. Copyright © 2014 Informa Healthcare USA, Inc

Safety of metformin in COPD

Table 1. Comparison of baseline characteristics in metformin and non-metformin groups Na

Metformin group (n = 51)

Non-metformin group (n = 79)

p-value

Age – years

130

70.5 ± 9.6

74.7 ± 9.6

0.016

Female sex – n (%)

130

18 (35)

29 (37)

0.511

130

17 (35)

11 (14)

0.016

58

40 (30–60)

50 (32–65)

0.580

130

20 (39)

38 (48)

0.368

Characteristic

Smoking history Current smoker – n (%) Pack-years Inhaled corticosteroid use – n (%)

COPD Downloaded from informahealthcare.com by Kainan University on 04/08/15 For personal use only.

Other treatments for diabetes – n (%) Insulin

130

6 (12)

22 (28)

0.031

Sulphonylurea

130

17 (33)

23 (29)

>0.99

Thiazolidinedione

130

3 (6)

0 (0)

0.070

Meglitinide

130

0 (0)

1 (1)

>0.99

Body weightb – kg

61

83.3 (71.4–104.1)

67.3 (52.2–79.8)

0.003

Male subjects

39

87.7 (79.2–108.8)

73.4 (64.5–88.0)

0.016

Female subjects

22

70.0 (63.4–97.0)

50.0 (39.4–64.0)

0.014

FEV1 – liters

32

0.93 (0.54–1.25)

0.92 (0.63–1.31)

0.833

FEV1 – % predicted

27

36 (23–48)

28 (11–45)

0.456

FVC – liters

32

1.78 (1.33–3.16)

2.06 (1.50–2.37)

0.803

Spirometry

Arterial blood gases at presentation pH

120

7.38 (7.33–7.43)

7.41 (7.33–7.44)

0.165

PO2 – kPa

120

8.60 (7.27–11.60)

8.71 (7.41–11.42)

0.703

PCO2 – kPa

120

6.33 (5.24–8.18)

5.94 (4.85–7.24)

0.194

Oxygen saturation – %

108

93 (90–96)

93 (91–97)

0.972

Hypoxia – n (%)

120

35 (75)

53 (73)

0.497

Respiratory failure – n (%)

120

20 (43)

25 (34)

0.234

Respiratory acidosis – n (%)

120

16 (34)

17 (23)

0.141

128

0.39 (0.36–0.44)

0.39 (0.36–0.44)

0.596

Creatinine – μmol/L

129

89 (75–114)

93 (79–144)

0.196

Estimated GFR – mL/min/1.73 m2

129

65 (53–87)

62 (39–79)

0.117

C reactive protein – mg/L

106

26 (6–72)

44 (14–95)

0.164

47

9.5 (6.8–13.3)

10.1 (7.1–14.5)

0.991

Hemoglobin A1c – %

51

8.2 (7.5–10.6)

7.8 (6.4–9.1)

0.140

Hemoglobin A1c – mmol/mol

51

66 (58–92)

62 (46–76)

0.140

Blood results Hematocrit

Random blood glucose – mmol/L

Charlson co-morbidity index

130

2 (2–3)

3 (2–4)

0.099

APACHE II score

130

13 (9–17)

13 (11–17)

0.621

Notes: aNumber of patients with data available. bHeight was recorded in too few patients to permit meaningful comparison of body mass index.

The median daily dosage of metformin was 1 (1–2) g. Among the 51 patients taking metformin at presentation, metformin was suspended during admission in 8 cases (16%) and stopped completely in 5 cases (10%). Among the 79 patients in the non-metformin group, metformin was started during inpatient stay in 4 cases (5%).

Lactate concentration The plasma lactate concentration was measured in 80 patients (62%): 36 (71%) in the metformin group and 44 (56%) in the non-metformin group. The median plasma www.copdjournal.com

lactate concentration was 1.45 mmol/L (1.10–2.05) in the metformin group and 1.10 mmol/L (0.80–1.50) in the non-metformin group (p = 0.012, Figure 1). The Hodges–Lehman estimate for the difference in medians was 0.40 mmol/L (95% confidence interval [CI] 0.10– 0.60). There were no cases of lactic acidosis (conventionally defined as lactate >5 mmol/L with academia and high anion gap (17)). Within the metformin group, there was no difference in the lactate concentration among patients in whom it was later stopped or suspended, as compared to those in whom it was continued (median 1.3 vs 1.5 mmol/L respectively, p = 0.410).

3

4

Hitchings et al.

not (1.9 years, 1.1–2.6; p = 0.011; Figure 2). This difference remained significant in a Cox regression model adjusting for the effects of age, sex, body weight, year of admission, co-morbidity burden (Charlson index), illness severity (APACHE-II score), hemoglobin A1c and blood glucose concentration (Table 2). Lactate concentration was not related to survival.

COPD Downloaded from informahealthcare.com by Kainan University on 04/08/15 For personal use only.

Discussion

Figure 1. First plasma lactate concentration at hospital admission. The horizontal bar indicates the median plasma lactate concentration (p = 0.012 Mann–Whitney test).

Survival Thirty-seven patients (28%) were still alive at the time vital status was ascertained; the median time to censoring was 8.1 years (IQR 5.1–10.7) after the index admission. Median survival overall was 3.3 years (95% CI 2.0–4.5). Survival was better for patients on metformin (median 5.2 years, 95% CI 4.5–5.8) than those

Safety of metformin We explored the safety of metformin in people with type 2 diabetes mellitus who had been admitted to hospital for COPD exacerbations. These patients were at high risk for lactate accumulation: approximately three-quarters were hypoxemic, over a third were in respiratory failure, and about a quarter had respiratory acidosis. Despite this, the lactate concentration was only slightly higher in patients taking metformin than in those not. Although statistically significant, the difference was small and fell below our predetermined, albeit subjective, threshold for clinical significance. As in non-COPD cohorts in which there have been similar findings (21, 22), this difference may be attributable to direct effects of metformin on mitochondrial function, or confounding by other variables such as body weight (22, 23). There were no cases of lactic acidosis and no evidence that metformin adversely affected survival. One prior study has assessed the safety of metformin in patients with traditional contraindications to its use

Figure 2. Kaplan-Meier curves for all-cause mortality. Patients on metformin are represented by a dashed line; those not on metformin by a solid line (p = 0.011, log-rank test). Copyright © 2014 Informa Healthcare USA, Inc

Safety of metformin in COPD

Table 2. Multivariate Cox proportional hazards regression model for death Hazard ratioa (95% CI)

Variable

p-value

Metformin therapy (present vs absent)

0.380 (0.150–0.967)

0.042

Age (years)

1.037 (0.980–1.097)

0.207

Sex (female vs male)

0.614 (0.234–1.608)

0.320

Year of admission

0.908 (0.779–1.059)

0.908

Body weight (kg)

1.027 (1.002–1.052)

0.034

Hemoglobin A1c (%)

1.008 (0.790–1.286)

0.951

Charlson co-morbidity index

1.186 (0.950–1.479)

0.131

0.906 (0.804–1.021)

0.105

b

APACHE-II score

COPD Downloaded from informahealthcare.com by Kainan University on 04/08/15 For personal use only.

Notes: aFor continuous variables, the hazard ratio represents the risk associated with a 1-unit increase. b Acute Physiology and Chronic Health Evaluation-II score.

(24). This included 91 patients with COPD, although those with a history of CO2 narcosis were excluded, and their exacerbation history was not described. Patients were randomized to stop or continue metformin, then followed for 4 years. There were no cases of lactic acidosis, and no difference in lactate concentrations between groups. Our study adds to this by providing data from a larger COPD cohort at greater risk of lactic acidosis, and with more lengthy survival data, albeit retrospective. The finding that survival was improved among patients taking metformin must be interpreted cautiously. The retrospective nature of the study introduces risks of biases and confounding, particularly as fitness may have factored into physicians’ assessments as to the appropriateness of metformin therapy. Of note, patients taking metformin were generally younger and had higher body weight than those not taking metformin (higher body weight is understood to confer a survival benefit in COPD (25)). That said, the prognostic significance of metformin was retained in a multivariate model adjusting for measurable confounders including age, body weight, and indices of acute illness severity and chronic co-morbidity burden. However, we cannot exclude effects from unmeasured or incompletely-captured confounders and biases. In this regard it is notable that we were unable to adjust for the severity of airflow obstruction due to limited spirometric data available.

Study limitations This study has several limitations, particularly in relation to its retrospective design. Decisions to undertake and record clinical measurements, and on the nature of patients’ treatment, all took place in the course of routine clinical practice. This introduces risks of confounding and bias as noted above. We relied on admission drug history to determine metformin exposure before admission, and on the discharge prescription to determine metformin exposure after hospital discharge. The resulting analyses therefore tend to suppose that all www.copdjournal.com

patients were fully adherent with their prescribed medicines, and that these did not change between hospital discharge and death. Inevitably, neither of these assumptions is correct. However, our study aimed to characterize the effect of metformin prescribing decisions in a real-world context, where subsequent management changes and incomplete adherence are to be expected; and in relation to mortality, the survival curves diverge early, when the discharge prescription is most likely to remain accurate.

Conclusion In summary, this retrospective study identified a small elevation in plasma lactate concentration among patients on metformin that was statistically significant but of doubtful clinical significance. There were no cases of lactic acidosis, and no evidence that metformin had an adverse effect on survival. The findings add to existing data that suggest COPD should not present a barrier to the use of metformin if clinically indicated. Moreover, in light of growing evidence from non-clinical and clinical studies that metformin may have additional beneficial properties of relevance in COPD, these data offer support for those who seek to investigate the effects of the drug in interventional trials. We have established one such trial, which is currently recruiting patients from eight centers in the United Kingdom (Current Controlled Trials ISRCTN66148745).

Acknowledgment We wish to acknowledge the significant and diligent contributions of Shaima Aslam, Richard Cull, Lawrence Hayes, Graham Picton and Luke Turner in collecting data used in this study.

Declaration of Interests Statement We have no relevant competing interests. The work described here had no specific funding. The clinical trial mentioned in the manuscript is supported by grants from the UK Medical Research Council and the British Lung Foundation. The authors are responsible for the content and the writing of this paper.

References 1. Wells CE, Baker EH. Metabolic syndrome and diabetes mellitus in COPD. In: Rabe KF, Wedzicha JA, Wouters EF, editors. European Respiratory Monograph COPD and Comorbidity. 59. Sheffield, UK: European Respiratory Society Journals Ltd., 2013. 2. Mannino DM, Thorn D, Swensen A, Holguin F. Prevalence and outcomes of diabetes, hypertension and cardiovascular disease in COPD. Eur Respir J 2008; 32(4):962–969. 3. Kinney GL, Black-Shinn JL, Wan ES, Make B, Regan E, Lutz S, et al. Pulmonary Function Reduction in Diabetes Mellitus with and without Chronic Obstructive Pulmonary Disease. Diabetes Care 2014; 37:389–395.

5

COPD Downloaded from informahealthcare.com by Kainan University on 04/08/15 For personal use only.

6

Hitchings et al. 4. American Diabetes Association. Standards of Medical Care in Diabetes—2013. Diabetes Care 2013; 36(Supplement 1):S11– S66. 5. National Collaborating Centre for Chronic Conditions (UK). Type 2 Diabetes: National Clinical Guideline for Management in Primary and Secondary Care (Update). London: Royal College of Physicians of London, 2008. Available from: http:// www.ncbi.nlm.nih.gov/books/NBK53885/. 6. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med 2008; 359(15):1577–1589. 7. Bailey CJ, Grant PJ, Evans M, De Fine Olivarius N, Andreasen A, Fowler P, et al. The UK prospective diabetes study. Lancet 1998; 352(9144):1932–1934. 8. Sexton P, Metcalf P, Kolbe J. Respiratory effects of insulin sensitisation with metformin: A prospective observational study. COPD: Journal of Chronic Obstructive Pulmonary Disease 2014; 11:133–142. 9. Kim HJ, Lee JY, Jung HS, Kim DK, Lee SM, Yim JJ, et al. The impact of insulin sensitisers on lung function in patients with chronic obstructive pulmonary disease and diabetes. Int J Tuberc Lung Dis 2010; 14(3):362–367. 10. Huang NL, Chiang SH, Hsueh CH, Liang YJ, Chen YJ, Lai LP. Metformin inhibits TNF-alpha-induced IkappaB kinase phosphorylation, IkappaB-alpha degradation and IL-6 production in endothelial cells through PI3K-dependent AMPK phosphorylation. Int J Cardiol 2009; 134(2):169– 175. 11. Isoda K, Young JL, Zirlik A, MacFarlane LA, Tsuboi N, Gerdes N, et al. Metformin inhibits proinflammatory responses and nuclear factor-kappaB in human vascular wall cells. Arterioscler Thromb Vasc Biol 2006; 26(3):611–617. 12. Myerburg MM, King JD, Jr., Oyster NM, Fitch AC, Magill A, Baty CJ, et al. AMPK agonists ameliorate sodium and fluid transport and inflammation in cystic fibrosis airway epithelial cells. Am J Respir Cell Mol Biol 2010; 42(6):676–684. 13. Formoso G, De Filippis EA, Michetti N, Di Fulvio P, Pandolfi A, Bucciarelli T, et al. Decreased in vivo oxidative stress and decreased platelet activation following metformin treatment in newly diagnosed type 2 diabetic subjects. Diabetes Metab Res Rev 2008; 24(3):231–237. 14. Bonnefont-Rousselot D, Raji B, Walrand S, Gardes-Albert M, Jore D, Legrand A, et al. An intracellular modulation of free radical production could contribute to the beneficial effects of metformin towards oxidative stress. Metabolism 2003; 52(5):586–589.

15. Garnett JP, Baker EH, Naik S, Lindsay JA, Knight GM, Gill S, et al. Metformin reduces airway glucose permeability and hyperglycaemia-induced Staphylococcus aureus load independently of effects on blood glucose. Thorax. 2013;68(9): 835–845. 16. British National Formulary (online). London: BMJ Group and Pharmaceutical Press; 2013. Available from: http://www.bnf. org/bnf/index.htm. 17. Bristol-Myers Squibb Company. Glucophage (metformin hydrochloride) tablets package insert, Princeton, NJ, USA, 2009. Available from: http://packageinserts.bms.com/pi/pi_ glucophage.pdf. 18. Salpeter SR, Greyber E, Pasternak GA, Salpeter EE. Risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus. Cochrane Database Syst Rev. 2010(4). doi: 10.1002/14651858.CD002967.pub4. 19. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40(5):373–383. 20. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13(10):818–829. 21. Mongraw-Chaffin ML, Matsushita K, Brancati FL, Astor BC, Coresh J, Crawford SO, et al. Diabetes medication use and blood lactate level among participants with type 2 diabetes: the atherosclerosis risk in communities carotid MRI study. PLoS ONE 2012; 7(12):e51237. 22. Davis TM, Jackson D, Davis WA, Bruce DG, Chubb P. The relationship between metformin therapy and the fasting plasma lactate in type 2 diabetes: The Fremantle Diabetes Study. Br J Clin Pharmacol 2001; 52(2):137–144. 23. Dykens JA, Jamieson J, Marroquin L, Nadanaciva S, Billis PA, Will Y. Biguanide-induced mitochondrial dysfunction yields increased lactate production and cytotoxicity of aerobicallypoised HepG2 cells and human hepatocytes in vitro. Toxicol Appl Pharmacol 2008; 233(2):203–210. 24. Rachmani R, Slavachevski I, Levi Z, Zadok B, Kedar Y, Ravid M. Metformin in patients with type 2 diabetes mellitus: reconsideration of traditional contraindications. Eur J Intern Med 2002; 13(7):428. 25. Hallin R, Gudmundsson G, Suppli Ulrik C, Nieminen MM, Gislason T, Lindberg E, et al. Nutritional status and long-term mortality in hospitalised patients with chronic obstructive pulmonary disease (COPD). Respir Med 2007; 101(9):1954– 1960.

Copyright © 2014 Informa Healthcare USA, Inc

Safety of Metformin in Patients with Chronic Obstructive Pulmonary Disease and Type 2 Diabetes Mellitus.

Type 2 diabetes mellitus (T2DM) is commonly associated with chronic obstructive pulmonary disease (COPD). Metformin is a valuable treatment for T2DM, ...
822KB Sizes 4 Downloads 5 Views