Comparative effectiveness of renin-angiotensin system blockers and other antihypertensive drugs in patients with diabetes: systematic review and bayesian network meta-analysis
BMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f6008 (Published 24 October 2013) Cite this as: BMJ 2013;347:f6008
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Rados and colleagues comment on the methods of our network meta-analysis (1), concerning about the issue of enrolling randomised clinical trials exclusively on diabetic subjects. They also highlight the potential beneficial effects of blood pressure reduction and thiazide diuretics, on which our meta-analysis did not mainly focus.
The pathogenesis of diabetic nephropathy is different from non-diabetic kidney diseases, and the choice of treatments may not be the same. The most recent guidelines suggest the use of either angiotensin converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) as the first line treatment for diabetic patients with hypertension (2, 3). As a result, the main objective of our study is to assess the difference between ACE inhibitors and ARBs, either in monotherapy or combination therapy, for their protective effects on patients with diabetes, and our findings are generally compatible with the recommendation of current guidelines (1).
We agree that blood pressure reduction also plays a renoprotective role in addition to the antiproteinuric effect of ACE inhibitors. However, various classes of antihypertensive treatments were associated with only small differences in blood pressure reduction, and the statistical power is low in metaregression or subgroup analysis for network meta-analysis simultaneously comparing many treatment strategies (4, 5). Therefore blood pressure reduction would only have a small impact on our results.
Only one trial assessing non-thiazide diuretics was enrolled in our network meta-analysis (6). This trial provided 27 participants who received the combination treatment of lisinopril and spironolactone and only reported all cause mortality (6); sensitivity analysis showed the same ranking order and similar estimates of effect size after omitting this trial (supplementary table E in our network meta-analysis) (1). Thus the estimated effects of diuretics on survival and major renal outcomes in our meta-analysis actually represent the effects from thiazide and thiazide-like diuretics.
About one third of the study participants in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) had type 2 diabetes, which contributed a substantial number of diabetic subjects (7). However, the ALLHAT study enrolled hypertensive patients with risk factors for heart attack, and did not randomise participants according to the status of diabetes (8). Similar conditions were also observed in other large randomised clinical trials, for example the HOPE study or the ACCOMPLISH study (9, 10). Including data of diabetic subpopulation from those trials can increase the sample size in the meta-analysis, but might also introduce additional heterogeneity and break the randomisation in the network. Because diabetic nephropathy is a major cause of end stage renal disease, our meta-analysis aims to specifically focus on diabetes and enroll trials exclusively on patients with diabetes.
Gatto comment on issues of zero event and low event rates that appeared in our meta-analysis (1), and shared experiences for handling those computational problems. Zero event is a common problem, when meta-analysis is used to pool rare events that are often analysed as secondary outcomes or adverse events in a trial. Without special statistical techniques of correction, such as the methods discussed by Gatto, data from trials with zero events in both arms will cause computational problem in traditional frequentist meta-analysis. The bayesian meta-analysis has the advantage to overcome the trials of zero events, and therefore the data of those trials can be put to good use. However, if the pooled participant number and event rate are too small in a specific treatment arm, uncertainty with wide credible interval can occur.
Sensitivity analysis by omitting trials with low event rates or rare participant numbers can be a practical way to assess the potential influences of those trials. We performed additional sensitivity analyses by omitting trials with zero events in all arms and treatment strategy with rare participants (ARB plus calcium channel blocker; ARB plus diuretic; ACE inhibitor plus ARB). The sensitivity analyses included 32 trials reporting all cause mortality, 11 trials reporting end stage renal disease, and 11 trials reporting the doubling of serum creatinine levels. Table 1 summarises the results of sensitivity analyses, which showed the same ranking orders and statistical significance, as well as slight changes in estimates of effect size and credible interval, compared with the results of our original analyses.
In conclusion, the results of our network meta-analysis are fairly robust, showing little change in sensitivity analyses, and therefore provide evidence for the clinical guidelines on antihypertensive management in patients with diabetes.
References
1. Wu HY, Huang JW, Lin HJ, Liao WC, Peng YS, Hung KY, et al. Comparative effectiveness of renin-angiotensin system blockers and other antihypertensive drugs in patients with diabetes: systematic review and bayesian network meta-analysis. BMJ 2013; 347: f6008.
2. Mancia G, Fagard R, Narkiewicz K, Redon J, Zanchetti A, Bohm M, et al. 2013 ESH/ESC guidelines for the management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). Eur Heart J 2013; 34: 2159-219.
3. American Diabetes Association. Standards of medical care in diabetes--2013. Diabetes Care 2013; 36 Suppl 1: S11-66.
4. Psaty BM, Lumley T, Furberg CD, Schellenbaum G, Pahor M, Alderman MH, et al. Health outcomes associated with various antihypertensive therapies used as first-line agents: a network meta-analysis. JAMA 2003; 289: 2534-44.
5. Sciarretta S, Palano F, Tocci G, Baldini R, Volpe M. Antihypertensive treatment and development of heart failure in hypertension: a Bayesian network meta-analysis of studies in patients with hypertension and high cardiovascular risk. Arch Intern Med 2011; 171: 384-94.
6. Mehdi UF, Adams-Huet B, Raskin P, Vega GL, Toto RD. Addition of angiotensin receptor blockade or mineralocorticoid antagonism to maximal angiotensin-converting enzyme inhibition in diabetic nephropathy. J Am Soc Nephrol 2009; 20: 2641-50.
7. Rahman M, Pressel S, Davis BR, Nwachuku C, Wright JT, Jr., Whelton PK, et al. Renal outcomes in high-risk hypertensive patients treated with an angiotensin-converting enzyme inhibitor or a calcium channel blocker vs a diuretic: a report from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Arch Intern Med 2005; 165: 936-46.
8. Davis BR, Cutler JA, Gordon DJ, Furberg CD, Wright JT, Jr., Cushman WC, et al. Rationale and design for the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). ALLHAT Research Group. Am J Hypertens 1996; 9: 342-60.
9. The HOPE study investigators. The HOPE (Heart Outcomes Prevention Evaluation) Study: the design of a large, simple randomized trial of an angiotensin-converting enzyme inhibitor (ramipril) and vitamin E in patients at high risk of cardiovascular events. Can J Cardiol 1996; 12: 127-37.
10. Jamerson KA, Bakris GL, Wun CC, Dahlof B, Lefkowitz M, Manfreda S, et al. Rationale and design of the avoiding cardiovascular events through combination therapy in patients living with systolic hypertension (ACCOMPLISH) trial: the first randomized controlled trial to compare the clinical outcome effects of first-line combination therapies in hypertension. Am J Hypertens 2004; 17: 793-801.
Competing interests: No competing interests
In the recent meta-analysis conducted by Wu et al. (1) the authors summarized data of trials that evaluated the effects of different classes of anti-hypertensives on survival and renal outcomes in patients with diabetes. As main results, beta-blockers were associated with negative effects on mortality, and ACE inhibitors were protective against doubling creatinine levels. Furthermore, the ACE inhibitor plus calcium channel blocker combination had the highest probability of being the best treatment, and diuretics (and their association with ACE inhibitors) were not better than placebo for any of the accessed outcomes.
Although interesting, these results should be interpreted with caution, considering the following issues.
First, as the authors acknowledge in their discussion, the results were not adjusted for the effects on blood pressure levels. The renoprotective effect of ACE inhibitors has been attributed to their unique antiproteinuric effect, but the intrinsic blood pressure reduction must be considered as shown by results of large direct comparative clinical trials (2). This fact limits conclusions about class effects, which was the main objective of the study.
Second, classifying all diuretics in the same group does not seem logical, as it is well recognized that there are diverse effects on blood pressure and outcomes of the distinct thiazide diuretics (3) and even among potassium-sparing diuretics.
Third, even though the authors selected a number of studies, which provided a well-connected network (at least in the mortality analysis), they included only studies conducted exclusively on diabetic subjects. This strategy led to the exclusion of data from large direct comparative studies, such as the ALLHAT (4), which included a wide number of diabetic patients in prespecified analyses that did not show any superiority of the ACE inhibitors over chlortalidone (5).
In conclusion, we believe these results should be interpreted with caution and the benefits of thiazide diuretics should not be ignored.
1) Wu HY, Huang JW, Lin HJ, Liao WC, Peng YS, Hung KY, Wu KD, Tu YK, Chien KL. Comparative effectiveness of renin-angiotensin system blockers and other antihypertensive drugs in patients with diabetes: systematic review and bayesian network meta-analysis. BMJ. 2013 Oct 24;347:f6008. doi: 10.1136/bmj.f6008.
2) Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugs in the prevention of cardiovascular disease: meta-analysis of 147 randomised trials in the context of expectations from prospective epidemiological studies. BMJ. 2009;338:b1665.
3) Roush GC, Holford TR, Guddati AK. Chlorthalidone compared with
hydrochlorothiazide in reducing cardiovascular events: systematic review and network meta-analyses. Hypertension. 2012 Jun;59(6):1110-7
4) ALLHAT group. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002 Dec 18;288(23):2981-97.
5) Rahman M, Pressel S, Davis BR, Nwachuku C, Wright JT Jr, Whelton PK, Barzilay J, Batuman V, Eckfeldt JH, Farber M, Henriquez M, Kopyt N, Louis GT, Saklayen M, Stanford C, Walworth C, Ward H, Wiegmann T. Renal outcomes in high-risk hypertensive patients treated with an angiotensin-converting enzyme inhibitor or a calcium channel blocker vs a diuretic: a report from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Arch Intern Med. 2005 Apr 25;165(8):936-46
Competing interests: No competing interests
In conducting their meta-analysis, Wu et al. (1) addressed the well-known issue of the problems caused by low event rates and zero events in meta-analysis. In particular, in their Methods section the authors have observed that "many of the included studies showed low event rates or even no events in one or both treatment arms" and that "studies with zero events do not cause computational problems with a bayesian approach as it usually does with traditional frequentist methods"; clearly, mortality was the end-point that, in this meta-analysis, was exposed to the highest risk of these computational problems.
One advantage of the work by Wu et al.(1) is that crude event frequencies were reported for individual studies. If the comparison of angiotensin-receptor blockers (ARB) vs placebo is considered according to the end-point of all-cause mortality, there were 11 direct-comparison studies reporting the event frequency for the ARB arm (data-set 1; 11 studies) and the controls (data-set 2; 11 studies). Individual crude rates were the following (first author, events/total in ARB arm and events/total in the controls): 1) IRMA-2, 3/389 and 1/201; 2) RENAAL, 158/751 and 155/762; 3) ABCD-2V,1/66 and 0/63; 4) DIRECT-Prevent1, 7/711 and 5/710; 5) DIRECT-Protect1, 7/951 and 8/954; 6) DIRECT-Protect2 , 37/951 and 35/954; 7) PERRIN et al , 0/7 and 0/6; 8) ROADMAP, 26/2232 and 15/2215; 9) RASS, 1/96 and 1/95; 10) MUIRHEAD et al, 0/62 and 0/31; 11) IDNT, 83/567 and 93/569 (data from Table 1 of reference 1; see reference 1 for the complete references of these 11 clinical studies).
The meta-analysis by Antoniou et al.(2), that was focused on the outcomes observed with carotid endartetectomy (CEA) or carotid artery stenting (CAS) in young and elderly patients, represents another recently published article in which study-specific crude rates of mortality were reported for different patient cohorts, namely: [a] old patients treated with CEA (data-set 3); [b] old patients treated with CAS (data-set 4); [c] young patients treated with CEA (data-set 5); [d] young patients treated with CAS (data-set 6). As regards the end-point of stroke, similar rates of incidence were reported by Antoniou et al. (2) for the same 4 patient cohorts, namely: [a] old patients treated with CEA (data-set 7); [b] old patients treated with CAS (data-set 8); [c] young patients treated with CEA (data-set 9); [d] young patients treated with CAS (data-set 10).
Although there was no specific criterion for selecting these 10 data-sets, this heterogeneous material was used to investigate the extent to which the pooled rates of proportion meta-analysis are affected by the computational approach adopted. For this purpose, we compared the following 4 computational approaches available in the Open Meta-Analysist Software (OMA)(3): A) no transformation of rates; B) log-transformation of rates; C) arcsine transformation of rates; and D) Freeman-Tukey transformation of rates. All of these approaches were based on the random-effect model and incorporated the 0.5 approximation that OMA introduces for replacing zero frequencies. The criterion for evaluating the results was simply given by the maximum percent deviation (highest vs lowest value) for each of the 10 data-sets assessed across the four methods.
Table 1 summarises the results of our analysis (in which many decimal digits were kept to better explore these computational problems). The performance indexes shown in the last column of Table 1 confirm that the (relative) differences between pooled rates estimated by different methods are wide (mean percent deviation for highest vs lowest = +42.1%; range +11% to +83%). Of course, the absolute differences across the four methods were small, but this simply reflects the low event rates typically found in these circumstances.
In conclusion, the present experience of comparative analysis between different computation methods applied to a series of critical datasets confirms that this problem exists, but, on the other hand, authoritative recommendations for managing this problem are still lacking.
References
1) Wu HY, Huang JW, Lin HJ, Liao WC, Peng YS, Hung KY, Wu KD, Tu YK, Chien KL. Comparative effectiveness of renin-angiotensin system blockers and other antihypertensive drugs in patients with diabetes: systematic review and bayesian network meta-analysis. BMJ. 2013 Oct 24;347:f6008. doi: 10.1136/bmj.f6008.
2) Antoniou GA, Georgiadis GS, Georgakarakos EI, Antoniou SA, Bessias N, Smyth JV, Murray D, Lazarides MK. Meta-analysis and meta-regression analysis of outcomes of carotid endarterectomy and stenting in the elderly. JAMA Surg. 2013 Oct 23. doi: 10.1001/jamasurg.2013.4135. [Epub ahead of print]
3) Open Meta-Analyst (OMA), software version 4.16.12, Tufts University, U.S., url http://tuftscaes.org/open_meta/).
Competing interests: No competing interests
We agree with the authors that renin-angiotensin system blockers (the ACE inhibitor and ARBs or angiotensin receptor blockers)are the best for hypertension in diabetic patients1. However, the diabetics are predisposed to atherosclerosis including atherosclerotic renal artery which if bilateral can cause renal failure due to decrease in perfusion pressure (and hence the glomerular filtration)to kidney as a result of lowering of hypertension which was maintaining the renal perfusion2. So, it is mandatory to record and investigate for bilateral renal artery stenosis due to atherosclerosis which is to be ruled out before administering renin-angiotensin system blockers3.
References:
1. BMJ 2013;347:f6008
2. Harrison's Principles of Internal Medicine, 17th edition, Volume 2. Publisher Mc GrawHill. P1555.
3. Main J . Atherosclerotic renal artery stenosis, ACE inhibitors, and avoiding cardiovascular death.Heart. 2005 April; 91(4): 548–552.doi: 10.1136/hrt.2003.019505
PMCID: PMC1768825
Competing interests: No competing interests
Re: Comparative effectiveness of renin-angiotensin system blockers and other antihypertensive drugs in patients with diabetes: systematic review and bayesian network meta-analysis
Reconfirming Recent Guidelines
Wu H-Y, et al’s paper1 is timely important in reconvincing physicans and patients the usefulness of updated guidelines to choose antihypertensive drugs2,3. They reported that compared with placebo, only angiotensin converting enzyme (ACE) inhibitors significantly reduced the doubling of serum creatinine levels and ACE inhibitor plus calcium channel blocker combination therapy had the greatest probability for being the best treatment on reducing mortality. Therefore, they concluded that the renoprotective effects and superiority of using ACE inhibitors in patients with diabetes, supporting the use of ACE inhibitors as the first line antihypertensive agent in patients with diabetes, considering the cost of drugs. Calcium channel blockers might be the preferred treatment in combination with ACE inhibitors if adequate blood pressure control cannot be achieved by ACE inhibitors alone.
Systemic hypertension and diabetes are associated with endothelial dysfunction that promotes inflammation, oxidation of lipoproteins, smooth muscle proliferation, extracellular matrix deposition or lysis, accumulation of lipid-rich material, platelet activation, thrombus formation, and insulin resistance. All of these consequences of endothelial dysfunction and insulin resistance may contribute to development and clinical expression of atherosclerosis.4,5
Chronic inflammation and oxidative stress play crucial roles in endothelial dysfunction, insulin resistance, and atherosclerosis.6 Chronic inflammation is a pathogenic feature of atherosclerosis and cardiovascular disease mediated by various substances including angiotensin II, proinflammatory cytokines, and free fatty acids. Moreover, additional mechanisms contributing independently to both insulin resistance and endothelial dysfunction include glucotoxicity, lipotoxicity, and inflammation together with oxidative stress. Cross-talk between inflammatory signaling pathways and insulin signaling pathways causes both metabolic insulin resistance and endothelial dysfunction that synergize to predispose to cardiovascular disorders, type 2 diabetes mellitus, metabolic syndrome, and cerebrovascular disease.7,8 Decreased production of nitric oxide mediated by endothelial dysfunction and insulin resistance contributes to accelerated atherosclerosis by multiple mechanisms. Controlling a variety of risk factors causing inflammation and oxidative stress with renin-angiotensin-aldosterone system blockade therapy may simultaneously address multiple mechanisms underlying the pathogenesis of atherosclerosis. This emerging therapeutic paradigm for slowing progression of atherosclerosis has advantages.9
In this regard, previous guidelines recommended to use ACE inhibitors to slow progression of renal damage and reduce cardiovascular morbidity and mortality in patients with type 2 diabetes mellitus.10 Further, 2013 European and American guidelines recommend ACE inhibitors in combination with calcium channel blockers as the preferred treatment because of a prompter response in a larger number of patients (potentially beneficial in high-risk patients), a greater probability of achieving the target blood pressure in patients with higher blood pressure values, and a lower probability of discouraging patient adherence with many treatment changes.2,3
Funding: None, Disclosures: None
REFERENCES
1. Wu HY, Huang JW, Lin HJ, Liao WC, Peng YS, Hung KY, et al. Comparative effectiveness of renin-angiotensin system blockers and other antihypertensive drugs in patients with diabetes: systematic review and bayesian network meta-analysis. BMJ 2013;347:f6008.
2. Mancia G, Fagard R, Narkiewicz K, Redón J, Zanchetti A, Böhm M, et al; Task Force Members. 2013 ESH/ESC Guidelines for the management of arterial hypertension: the Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2013;31:1281-357.
3. Go AS, Bauman M, Coleman King SM, Fonarow GC, Lawrence W, Williams KA, et al. AHA/ACC/CDC Science Advisory: An Effective Approach to High Blood Pressure Control A Science Advisory From the American Heart Association, the American College of Cardiology, and the Centers for Disease Control and Prevention. J Am Coll Cardiol 2013 Nov 12. doi:pii: S0735-1097(13)06077-4. 10.1016/j.jacc.2013.11.007.
4. Koh KK. Effects of statins on vascular wall: vasomotor function, inflammation, and plaque stability. Cardiovasc Res 2000;47:648-57.
5. Koh KK, Han SH, Quon MJ. Inflammatory markers and the metabolic syndrome: insights from therapeutic interventions. J Am Coll Cardiol 2005;46:1978-85.
6. Koh KK, Oh PC, Quon MJ. Does reversal of oxidative stress and inflammation provide vascular protection? Cardiovasc Res 2009;81:649-59.
7. Han SH, Quon MJ, Koh KK. Reciprocal relationships between abnormal metabolic parameters and endothelial dysfunction. Curr Op Lipidol 2007;18:58-65.
8. Muniyappa R, Montagnani M, Koh KK, Quon MJ. Cardiovascular actions of insulin.
Endocr Rev 2007;28:463-91.
9. Lee H-Y, Sakuma I, Ihm S-H, Goh C-W, Koh KK. Statins and renin-angiotensin system inhibitor combination treatment to prevent cardiovascular disease. Cir J 2014 (in press)
10. Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G, et al; ESH-ESC Task Force on the Management of Arterial Hypertension. 2007 ESH-ESC Practice Guidelines for the Management of Arterial Hypertension: ESH-ESC Task Force on the Management of Arterial Hypertension. J Hypertens 2007;25:1751-62.
Competing interests: No competing interests