Potential benefits of statins on morbidity and mortality in chronic obstructive pulmonary disease: a review of the evidence

Transcription

1 1 Department of Mediine, University of Aukland, Aukland; 2 Respiratory Servies, Aukland City Hospital, Aukland, New Zealand Correspondene to: Dr R Young, Department of Mediine, Aukland Hospital, Private Bag 92019, Aukland, New Zealand; roberty@ adhb.govt.nz Reeived 4 January 2009 Aepted 28 Marh 2009 Potential benefits of statins on morbidity and mortality in hroni obstrutive pulmonary disease: a review of the evidene R P Young, 1 R Hopkins, 1 T E Eaton 2 ABSTRACT Studies show redued fored expiratory volume in 1 s (FEV 1 ) in patients with hroni obstrutive pulmonary disease (COPD) is an important independent preditor of ardiovasular death and is haraterised by both pulmonary and systemi inflammation. Evidene shows statins have important anti-inflammatory effets in both the lungs and arteries. Although randomised ontrol trials are yet to be reported, non-randomised studies have onsistently shown benefit in COPD patients taking statins ompared with those not. These inlude redutions in both ardiovasular and respiratory morbidity/mortality. Other potential benefits inlude a redued deline in FEV 1 and redued risk of lung aner. It is argued that onfounding by a healthy user effet is unlikely to explain the observed benefit. Given the undisputed benefit of statins in high risk populations and the growing body of data suggesting statins may benefit patients with COPD, the question arises Should statins be onsidered more often in patients with COPD?. Patients with hroni obstrutive pulmonary disease (COPD) die primarily from ompliations of smoking, speifially oronary artery disease (CAD), COPD related ompliations (respiratory failure with or without hest infetion), lung aner and stroke. 1 5 Colletively these aount for over 80% of deaths in COPD. 1 7 CAD is the most ommon ause of death, estimated to affet between 20 50%, 1 5 but both lung aner and COPD related ompliations are also ommon, estimated at 20 30% and 10 30% of deaths, respetively. 1 6 COPD and redued fored expiratory volume in 1 s (FEV 1 ) are both powerful preditors of mortality. Studies show that the severity of COPD, based on FEV 1, predits survival with an estimated mortality of 50% over 5 years. 389 This is omparable to mortality from advaned CAD and many forms of aner. Nevertheless, to date only smoking essation, long term oxygen, and lung redution surgery have shown benefit in this regard. 10 Bronhodilating and anti-inflammatory treatment for COPD have been shown to have only a limited effet on deline in lung funtion 11 and limited redution on survival Despite being the fourth leading ause of death in developed ountries, COPD is signifiantly under diagnosed and predited to be the third leading ause in the oming years. Urgent efforts are required to address this major publi health problem. LUNG FUNCTION AND CARDIOVASCULAR MORTALITY Redued FEV 1 is a powerful marker for CAD and mortality from ardiovasular disease after ontrolling for several potential onfounders. 23 Although smoking is impliated in both COPD and CAD, it is those smokers with poor lung funtion, estimated to be 20 30% of all smokers, who are at greatest risk of a oronary death. 3 6 Indeed for men, the ombination of redued FEV 1 and smoking exposure are better preditors of future mortality from heart disease than serum holesterol values. 10 In this study, where FEV 1 was ompared with traditional risk fators, it is striking that redued FEV 1 ranks seond only to smoking, well above blood pressure, soial lass and holesterol as a preditor for all ause mortality in both men and women. Further support for the importane of FEV 1 in ardiovasular risk assessment omes from two studies showing that patients with COPD have a higher prevalene of oexisting ardiovasular risk fators than those with normal lung funtion ; speifially, those with COPD had a higher prevalene (and presumed risk) for diabetes, hypertension and ardiovasular disease. 26 This study showed that by inluding COPD (based on spirometry) with traditional risk fators, further refinement of ardiovasular risk ould be ahieved. Although the seond study showed that the relationship between lung funtion and ardiovasular death was redued after adjustment for traditional risk fators, 27 COPD remained an independent and signifiant risk fator in its own right. This observation is onsistent with the hypothesis that the tendeny to exaggerated systemi inflammation underlies both COPD and ardiovasular disease (disussed further below). 28 Given the inreased CAD risk assoiated with COPD (or redued FEV 1 ), it seems very reasonable to onsider drug treatment to address this risk speifially. On this basis alone, it ould be argued that HMG CoA redutase inhibitors (statins) would benefit many patients with COPD for primary prevention of CAD. To exemplify the extent to whih lung funtion is relevant to ardiovasular mortality, the authors highlight data from several studies In these studies, redued FEV 1 is shown to be more important than smoking exposure after adjustment for other variables (fig 1), not only in ardiovasular mortality but in death from all auses. Among smokers of omparable smoking exposure, redued FEV 1 was assoiated with as muh as a 3 4 fold greater ardiovasular mortality Remarkably, this effet also extends to nonsmokers where poor FEV 1 predits a risk 2 3 fold greater than that of heavy smokers with normal lung funtion. We onlude that FEV 1 is an Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. 414 Postgrad Med J 2009;85: doi: /pgmj

2 Figure 1 Relationship between fored expiratory volume in 1 s (FEV 1 ), smoking status and (A) odds ratio for ardiovasular mortality (modified from Tokman et al 21 ) and (B) all ause mortality rate (%) (modified from Olofson et al 22 ). important marker of future ardiovasular mortality and that its effets on mortality are both independent of and synergisti with those from smoking. The underlying mehanism for this may relate to systemi inflammation whih has been impliated in COPD and ardiovasular disease. COPD AND LUNG CANCER Although smoking exposure has a entral role in both, only approximately 10 15% of hroni smokers develop lung aner while 20 30% develop COPD. Epidemiologial studies show that smokers with COPD are at onsiderably higher risk of lung aner than smokers with normal lung funtion, ontributing over 50% of all lung aner ases. There is growing evidene that COPD and lung aner result from ommon pathologial responses to inflammatory proesses in the lung, and that the individual smoker s response to these proesses are genetially determined Colletively, these studies show that an overlapping subgroup of smokers (and ex-smokers), haraterised by redued FEV 1, are at inreased risk of CAD, COPD and lung aner. COPD AND SYSTEMIC INFLAMMATION Reent studies have shown a onsistent assoiation between biomarkers of systemi inflammation, primarily C reative protein (CRP), and severity of COPD In a population based study (NHANES III), those with severe airways obstrution (FEV 1 % predited,50%) were twie as likely to have an elevated CRP value. 43 This study also reports an additive effet between the presene of moderate severe COPD and elevated CRP on risk of ardia injury. 43 The finding of an assoiation between systemi inflammation in both moderate and severe COPD support epidemiologial data showing that even small redutions in FEV 1 an inrease ardia morbidity and mortality 2 3 fold in the general ommunity Surprisingly, this relationship with CRP exists aross smokers, ex-smokers and non-smokers, although one small population study failed to identify any assoiation. 50 Two studies have examined the preditive utility of CRP in outome for patients with COPD, although the findings were inonlusive. There is ertainly a greater tendeny to onsider COPD a systemi disease where weight loss predates lung funtion deline 53 and non-pulmonary manifestations inlude diseases suh as anaemia and osteoporosis. In summary, there is growing evidene that COPD is assoiated with systemi inflammation that is exaggerated by, but not dependent on, igarette smoking. It is possible other aero-pollutants are important here 54 and that this suseptibility to systemi inflammation may be in part genetially onferred. These observations would be onsistent with the hypothesis that lung funtion is not just a barometer to the lung s response to airway aero-pollutant exposure but also a marker of a more general systemi response. 39 If this were true, then drug treatment with systemi anti-inflammatory benefits may be benefiial in reduing other inflammatory based diseases suh as CAD, lung aner or stroke. Interestingly, in a population study from Norway, statin use was assoiated with a derease in CRP onentration. 57 More importantly, statin use in a reently reported randomised ontrol trial of healthy people with elevated CRP and normal lipid profile reported redutions of 40 60% in ardiovasular end points 58 If systemi inflammation is onsidered an important determinant of mortality in COPD and a target for preventive treatment (as it is in CAD), then statins may have signifiant additional benefits in COPD where elevated CRP is a feature and existing treatments target pulmonary inflammation only. 59 In this regard, statin treatment may represent a new and muh needed adjunt treatment in the management of COPD. STATINS AND THEIR ANTI-INFLAMMATORY EFFECTS ON ARTERIES AND THE LUNGS In the management of oronary artery disease, statins play a entral role by signifiantly reduing the serum holesterol onentration. However, their role in redution of mortality and morbidity in CAD is thought to relate as muh to their antiinflammatory effets as to their holesterol lowering effets (that is, pleiotropy). 60 Studies in both primary and seondary prevention show that statin use is assoiated with a relative 30% redution in mortality (table 1). 61 The absolute benefit of statin treatment in seondary prevention (in terms of absolute risk redution) is muh greater and thus more ost effetive than in those at lower risk. 62 To maximise the benefit of statin therapy, treatment is targeted at patients who have established CAD or those who have a very high baseline risk of having a heart attak. People in the latter group are best identified by assessing their risk based on several linial parameters suh as age, holesterol value, blood pressure and family history (for example, Framingham sore). There is now onsiderable evidene suggesting that statins have immunomodulatory effets that ould attenuate the inflammatory effets of smoking on the lung, not just the arteries. These inlude reduing neutrophil migration, ytokine prodution, adverse matrix remodelling, small airways inflammation, and apoptosis Of note, neutrophil migration into the mouse lung following lipopolysaharide (LPS) indued inflammation is inhibited by statins. 68 As neutrophil mediated inflammation is entral to smoking effets on the lung in COPD, this effet of statins may be additive to those of inhaled ortiosteroids Persistene of neutrophils in COPD due to inhibition of neutrophil apoptosis and/or phagoyti learane might also be relevant to attenuating persisting Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. Postgrad Med J 2009;85: doi: /pgmj

3 Table 1 Risk redution assoiated with statin therapy in the observational studies ompared with primary prevention randomised ontrolled trials (RCTs) in oronary artery disease Cohort No OR (95% CI) Relative risk redution Referene COPD (0.42 to 0.62) 47% risk redution for death or MI in high risk COPD patients COPD (0.38 to 0.87) 43% risk redution for death after hospital disharge for COPD exaerbation COPD (0.43 to 0.91) 38% risk redution for death following hospitalisation with hest infetion 0.19 (0.08 to 0.47) 81% risk redution for death following hospitalisation with COPD exaerbation US Veterans (0.42 to 0.48) 55% risk redution of lung aner with. 6 months of statin use neutrophili inflammation. 68 We suggest that identifying smokers with signifiant air flow limitation (%predited FEV 1,70%) ould be onsidered analogous (in terms of future ardiovasular risk) to identifying those with elevated holesterol. 739 Evidene of airflow limitation, together with other CAD risk fators (for example, smoking, elevated blood pressure or family history) should prompt onsideration of primary preventive treatment with statin therapy in those with multiple risk fators. Given the substantial risk of CAD onferred by a redued FEV 1, it would not be surprising that patients with COPD might gain greater ardiovasular benefit from statin therapy than those with normal lung funtion from the primary prevention studies. STATINS IN COPD: FINDINGS FROM OBSERVATIONAL STUDIES Given the importane of redued FEV 1 in ardiovasular risk, we support others who propose that FEV 1 be used in onjuntion with ommonly used risk markers, suh as blood pressure and serum holesterol, to assess risk and target preventive treatment. Suh an approah may improve outomes as suggested by three reent large prospetive observational studies reporting that patients with COPD on statins had substantial redutions in both morbidity and mortality ompared with those with COPD who were not Strikingly, those with COPD who took statins had as muh as a 50% redution in all ause mortality, % redution in myoardial infartion, 70 and 30% redution in hospitalisation from COPD (table 1). 70 These findings were repliated in another observational study of patients with peripheral vasular disease that, after a median 5 year follow-up, showed those with COPD taking statins had mortality redution of 30 40% ompared with non-users. 73 In a reently reported non-randomised study of lung funtion sreening in smokers and ex-smokers with COPD (n = 319), ompared with those not on statins, those taking statins had a signifiantly redued annual FEV 1 deline (+5 ml/year ompared with 286 ml/year in those with mild COPD not taking a statin) and 37% redution in COPD related hospitalisation. 74 This is onsistent with other studies purporting a ml/year loss in Manini et al 70 Soyseth et al 71 Frost et al 72 Farwell et al 83 US Veterans (0.60 to 0.81) 30% risk redution of lung aner Karp et al 84 US Veterans (0.42 to 0.70) 46% risk redution of death following hospitalisation for pnuemonia Primary prevention RCT* Mortensen et al (0.56 to 0.91) 29% risk redution for death following an MI Pignone et al (0.62 to 0.79) 30% risk redution for having a MI CI, onfidene interval; COPD, hroni obstrutive pulmonary disease; MI, myoardial infartion; OR odds ratio. *Based on a meta-analysis of randomised ontrolled trials. FEV 1 in smokers most suseptible to COPD, versus the normal annual deline of approximately ml We note that the deline of FEV 1 in linial trial patients, with mild to moderate COPD, is estimated to be in the order of 50 ml/year As a randomised linial trial is needed to onfirm that statins attenuate FEV 1 deline, studies involving mild to moderate COPD might require greater numbers (to improve power) than those reported in these observational studies. The observation that statins might attenuate FEV 1 deline 74 is onsistent with data reported in a subpopulation of the Normative Aging Study involving 803 men and 2136 measurements over a 10 year followup period. 79 Aross a wide range of baseline lung funtion, the average yearly deline in FEV 1 was 24 ml/year in those not on statins and 11 ml/year in those on statins. Importantly, in both studies this effet was found regardless of smoking status, suggesting statin therapy benefits both smokers and ex-smokers When the lung funtion findings from these studies are extrapolated over a 20 year period, the preservation of lung funtion in a COPD patient taking statin therapy would be omparable to that ahieved from quitting smoking (fig 2). Although data from randomised studies are needed to onfirm these interesting findings, they suggest that statins ould be one of the first pharmaologial agents to preserve lung funtion, to date the only intervention other than smoking essation to do so. Although smoking essation is and must remain the priority in all patients, the addition of statins to urrent treatment appear to onfer protetive effets on the lung and may be useful in those ex-smokers with COPD where lung funtion ontinues to deline. These observational data are remarkably onsistent and suggest the mortality benefit in COPD patients taking statins was approximately twofold greater than the mortality benefit seen with inhaled ortiosteroid treatment (up to 25% redution) 80 or ortiosteroid treatment ombined with long ating bronhodilators (up to 35% redution). 81 Moreover, the redution in hospitalisation with the statin treatment was omparable to that ahieved with loally ating inhaled ortiosteroids, onsidered by many as routine therapy. 81 The apparent benefit of statins over Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. 416 Postgrad Med J 2009;85: doi: /pgmj

4 Figure 2 Estimated effet of statin treatment on fored expiratory volume in 1 s (FEV 1 ) deline in non-randomised prospetive studies. COPD, hroni obstrutive pulmonary disease. ortiosteroids may reflet their greater attenuating effet on neutrophili inflammation and/or apoptosis, both of whih are key features of COPD Perhaps of greater signifiane was that the redution in myoardial infartion seen with statin treatment was twofold greater in patients with COPD than was seen in the oronary artery primary prevention studies. 61 Lastly, statin use has also been assoiated in three large observational studies with redued risk of lung aner (table 1) Two US studies reported data from the Veterans Affairs Healthare System involving patients (from south Central VA) and (from New England). In the former ase ontrol study, statin use for.6 months was assoiated with a 55% redution in lung aner risk ompared with those not taking statins. 82 This redution was seen aross all ages and independent of other ardiovasular risk fators suh as smoking and the presene of diabetes. In the New England study, with a median 5 year follow up, there was a 30% risk redution for lung aner after adjustment for onfounding variables. 83 The third study reported data on patients and reported a 35 47% redution in the inidene of lung aner after 7 years of follow up. 84 In another Veteran Affairs study, statin use was assoiated with a 50% redution in 30 day mortality in those hospitalised with ommunity aquired pneumonia, 85 although the benefit of statin use in sepsis remains ontroversial Just as was seen in the COPD related studies above, there appears onsisteny in both the diretion and magnitude of effet with statin use and redued lung aner risk. Although these studies onsistently report benefit over harm, with redutions in statin users ranging between 30 50%, history shows that the real magnitude of the benefit of statin use may be less and best estimated from randomised linial trials. This is beause all these statin studies have been observational studies, where the potential for onfounding exists (see healthy user effet below). STATINS: FINDINGS FROM AN INTERVENTIONAL STUDY The Heart Protetion Study (HPS) was a large randomised linial trial of seondary prevention in patients with vasular disease or diabetes. 89 It onluded that statin users had a 17% redution in vasular deaths although no signifiant effet on non-vasular death was found. Over the 5 year period of the study no inrease in mortality from diseases suh as aner (and speifially lung aner) was observed, but it should be noted that this was a relatively young ohort to be exploring effets on aner inidene. In a sub-analysis, there was a nonsignifiant trend to redued death from all respiratory diseases and hospital admission for COPD. Findings showed death from all respiratory diseases were redued in statin users ompared with non-users (26 (0.3%) simvastatin vs 39 (0.4%) plaebo, for respiratory death relative redution of 34%, relative risk (RR) 0.66, 95% onfidene interval (CI) 0.41 to 1.08, p = 0.10; and 88 (0.9%) vs 110 (1.1%) for COPD admission relative redution of 21%, RR 0.79, 95% CI 0.60 to 1.05, p = 0.10). This was a small subsample from a large RCT, where there was no seletion (or sub-analysis) for patients with COPD. In this setting dilution effets may be large and result in under-powering for any effet between statins and respiratory morbidity/mortality. No signifiant differene in absolute FEV 1 was observed on the final visit in a subgroup taking simvastatin ompared with plaebo. It is diffiult to interpret this last finding as no lung funtion was done at baseline, smoking prevalene was low, and it is not lear whether the two groups were mathed for relevant fators suh as smoking exposure. As approximately 70 80% of smokers maintain normal or near normal lung funtion, 30 any benefiial effet of statin treatment over a 5 year period may be lost (or diluted) by those unlikely to benefit. The latter group will be large and omprise non-smokers (estimated at 80% of the ohort) and smokers who have maintained normal or near normal lung funtion despite smoking (estimated as 15% of the ohort). Thus, an estimated 95% of the HPS ohort, where little statin effet on lung funtion would be expeted, might obsure the benefiial effet of statin use in those who were smokers and had COPD (remaining 5%). This is beause the positive effets statins might onfer on the lungs ome from mitigating the inflammatory effets from smoking (see effets of statins on the lungs). Any benefit would thus be found primarily in those smokers where the inflammatory effets of smoking are exaggerated or maladaptive (that is, in suseptible smokers with COPD), leading to remodelling in both the airways and/or lung parenhyma. THE HEALTHY USER EFFECT The observation that patients with COPD taking statins do better than those not taking statins has been attributed to the healthy user effet. 90 This effet means any benefits assoiated with taking statins is atually seondary to other fators suh as o-existing disease, underlying COPD severity, other lifestyle fators (for example, smoking), drug ompliane and fators related to health are quality or aess. In other words, those taking statins are doing other things that improve their health status (for example, attending the dotor more, smoking less, are otherwise healthier with better lung funtion or generally are more ompliant with their mediations). Aordingly, improved outomes in statin users are not due to taking statins per se but some other unreognised fator assoiated with being presribed statins (that is, statin use is onfounding the observed better outomes). Although the healthy user effet may explain the mixed results from statin use in reduing mortality in infetion (sepsis or pneumonia), to attribute this effet to the onsistently reported mortality and morbidity benefits of statin use in the COPD studies outlined above requires detailed analysis. The benefit of statin use on mortality is greatest in COPD patients with pre-existing ardiovasular disease 70 and may relate to effets on reduing Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. Postgrad Med J 2009;85: doi: /pgmj

5 systemi inflammatory ytokines linked to oronary disease (for example, interleukin 6 (IL6)). 88 Suh an effet may also be of importane in the lungs and we note with interest the onsistently reported statin effet on reduing FEV 1 deline and reduing lung aner risk where pulmonary inflammation is likely to be most relevant First, patients taking statins have been reognised to have established CAD or risk fators for CAD in ontrast to those nonstatin users who presumably have not. In the New England Veterans study of partiipants, statin users had signifiantly higher diabetes prevalene, smoking prevalene, low density lipoprotein (LDL) values, lung diseases, and ardiovasular disease. 83 Only hypertension and mental illness were marginally higher in the non-statin user group ompared with the statin users. We onlude that baseline health status, and thus expeted mortality, appears worse in statin users ompared with non-users. Given how well statins are tolerated, 91 the proportion of patients not taking statins beause they do not tolerate them is likely to be small and non-ontributory. Seond, although those taking statins have been shown to seek out other health preventive servies 92 suh as influenza vaination, pneumooal vaination or aner sreening, there are several reasons why suh ativities might not explain a 50% redution in mortality. Only the minority of the population undertake these ativities (influenza vaine 38%, pneumooal vaine 6%, mammography 21% of women, and prostate speifi antigen (PSA) testing 21% of men). Furthermore, the magnitude of the differene was small (the relative frequeny of influenza vaination use among statin users was only 21% higher than in those not taking statins with an absolute differene of 10%). Given that the differene in uptake of these preventive therapies is modest and that fewer than half of statin users atually use these treatments, it would require a substantial benefit on mortality for any of these to aount for a 50% redution in mortality. To date, none of these preventive therapies have been onviningly shown to redue ardiorespiratory mortality or morbidity and therefore are unlikely to aount for a onfounding effet from statin use. Similarly, if statin use was assoiated with better drug ompliane with COPD related treatments suh as inhaled ortiosteroid use, the latter is unlikely to aount for a 50% redution in mortality aross all those with COPD. In one observational study, where the effet of a statin was ompared in those taking or not taking ortiosteroids, little further risk redution ame from steroid Table 2 Statin use and lung funtion in a tertiary entre hroni obstrutive pulmonary disease lini Charateristi Statin user (n = 125) Statin non-user (n = 137) Mean age (years) Male gender (%) 62% 60% Pak years Current smoker 34% 44% Lung funtion FEV 1 (litres) %predited FEV 1 67% 65% FEV 1 /FVC 59% 58% Past medial history Cardiovasular event (%)* 45% 8% Pneumonia (%) 37% 38% COPD exaerbation 41% 39% COPD, hroni obstrutive pulmonary disease; FEV 1, fored expiratory volume in 1 s; FVC, fored vital apaity. *p,0.05. Key referenes Young RP, Hopkins RJ, Eaton TE. FEV1 not just a lung funtion test but a marker of premature death from all auses. Eur Respir J 2007;30: MClay JD, MAllister DA, MaNee W. Cardiovasular risk in hroni obstrutive pulmonary disease. Respirology 2007;12: Rodriguz-Roisin R, Soriano JB. Chroni obstrutive pulmonary disease with lung aner and/or ardiovasular disease. Pro Am Thora So 2008;5: Hothersall E, MSharry C, Thomson NC. Potential therapeuti role for statins in respiratory disease. Thorax 2006;61: Cazzola M, Ciaprini C, Page CP, et al. Targeting systemi inflammation: novel therapies for treatment of hroni obstrutive pulmonary disease. Expert Opin Ther Targets 2007;11: use. 70 It is not apparent what other treatments the statin users were taking whih might onfer this degree of effet. Third, both population and observational studies show there are muh higher rates of ardiovasular disease in those with COPD taking statins ompared with non-users As there is no evidene that having COPD inreases your lipid onentrations, it would seem likely that statin users are inherently at greater risk of ardiovasular disease (with greater expeted mortality) than those not presribed them. One of the observational studies attempted to look at this by subdividing their COPD ohort aording to baseline CAD risk 70 and found signifiantly redued rates of COPD related hospitalisation and deaths in the low risk group, although myoardial infartion redution was no longer evident. Similarly, the benefits of taking inhaled ortiosteroids did not ontribute to better outomes in this study. 70 Another fator assoiated with statin use was a higher frequeny of sreening for aner. 92 Sreening for aner may redue aner mortality through early diagnosis and treatment, but does not redue aner prevalene. In the previously disussed observational study from New England, the lung aner prevalene was lower in statin users ompared with nonusers. 82 This is unlikely to be due to a greater prolivity for, or awareness of, sreening in lung aner. Several of the observational studies have ontrolled for other lifestyle fators suh as smoking and still find a mortality benefit with statin use. Protetive lifestyle fators suh as intake of fresh fruits and vegetables may be higher in statin users ompared with non-users, but again the proportion of people involved and the magnitude of the effet are likely to be too small to aount for the mortality benefit seen in the observational studies. One possibility is that the ombined effets of all these potential onfounding variables might explain the benefit assoiated with statin use; however, this would suggest these benefiial onfounders were relevant aross all populations studied by these observational studies. Lastly, there are only limited data on just how frequently statins are presribed in patients with COPD and no evidene to suggest that statin use is more ommon in those with milder forms of COPD (that is, onfounding by severity of COPD). Data from observational studies suggest that the proportion of patients with COPD presribed statins to be on average 26% (range 20 50%). This orrelates with other population based studies that report 29% of patients diagnosed with COPD Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. 418 Postgrad Med J 2009;85: doi: /pgmj

6 Key learning points Redued fored expiratory volume in 1 s (FEV 1 ) is assoiated with systemi inflammation and is an independent marker of ardiovasular morbidity and mortality. Redued FEV 1 is also an independent marker for risk of lung aner and stroke. Statins have anti-inflammatory effets in both arteries and airways through attenuation of neutrophili inflammation. Reported benefits of statin use in patients with hroni obstrutive pulmonary disease (COPD) inlude redued ardiovasular mortality, redued respiratory morbidity/ mortality, redued lung funtion deline, and redued risk of lung aner. Although these benefits have yet to be onfirmed in randomised ontrolled trials, they are unlikely to be explained by the healthy user effet. were taking statins 84 and 34% with lung disease. 83 We examined this variable in a small audit of our COPD lini patients and found no differene in demographi variables or lung funtion between those taking and not taking statins (table 2). However, among the 48% (n = 125) who were urrently taking statins, nearly a half (45%, 56/125) reported having had a ardiovasular event (angina, myoardial infartion, oronary revasularisation or stroke) ompared with 8% (11/137) in the non-statin group. Similar findings have been reported by others. Given the evidene to date, it seems the healthy user effet is very unlikely to explain the onsistent redutions in mortality and morbidity in those taking statins in these large observational studies. In the USA alone there are over 90 million smokers or ex-smokers and an estimated 15 million with COPD who might benefit from the use of statin treatment as a preventive agent for not just CAD but COPD and lung aner as well. Colletively these diseases aount for 70 80% of premature death in smokers. 93 Statins are urrently one of the most widely presribed drugs and are generally well tolerated with a safe side effet profile. Analogous to the linial situation in CAD, there ould be targeting of smokers and exsmokers for preventive statin treatment using a risk marker suh as redued FEV 1. We onlude from these data that although it is possible that an as yet unidentified onfounding fator assoiated with statin use might explain a 50% mortality redution, to date, no onvining explanation has been reported. SUMMARY Based on growing epidemiologial, laboratory and linial studies, there is now onsiderable evidene to suggest that statin treatment ould improve outomes in COPD. This ould have onsiderable impat on this important disease. Given the importane of redued FEV 1 in prediting ardiovasular disease, we support others who propose that it be used in onjuntion with aepted risk markers, suh as blood pressure, CRP and serum holesterol, to assess risk and target preventive treatment. Targeted treatment ould also be based on more sophistiated models suh as the BODE index. 96 Suh an approah may improve outomes in the most ost effetive manner. The magnitude of this respiratory benefit may be as muh as twofold greater than that from existing treatments (for example, inhaled ortiosteroids), while the magnitude of the ardiovasular benefit may be nearly twofold greater than that Current researh questions In a randomised ontrol trial, how does the anti-inflammatory benefit of statin treatment ompare with inhaled ortiosteroids? Will randomised ontrol trials repliate the diverse morbidity and mortality benefits of statin treatment onsistently reported in observational studies? urrently seen in heart attak prevention where mortality has halved sine the introdution of statins. Although it is possible that this statin benefit is onfounded by other fators independently assoiated with statin use (the so alled healthy user effet ), there is no good evidene to suggest this is the ase. The observational studies typially show that statin users tend to be older and have more o-morbidities (suh as oronary artery disease, dyslipidaemia and diabetes) than non-users that is, not healthier than non-users. There is no apparent differene in lung funtion between those presribed and not presribed statins. Moreover, the preventive fators reported to be assoiated with statin use appear insuffiient in frequeny and effet size to aount for the morbidity and mortality benefits seen with statin use. We and others have previously reported that redued FEV 1 is an independent preditor of mortality from both ardiovasular disease (CAD and stroke) and respiratory disease (COPD and lung aner) We have also proposed that redued FEV 1 provides a non-invasive marker of suseptibility to aero-pollutant exposure and tendeny to systemi inflammation. 39 Here we suggest that patients with COPD be onsidered for statin treatment for three reasons. Firstly, a diagnosis of COPD (or redued FEV 1 ) is in its own right a powerful marker of inreased risk of oronary artery disease. 7 8 This risk is greater than that of other well established risk fators usually prompting statin treatment (elevated holesterol, smoking, family history, diabetes and elevated blood pressure). Seond, there is onsisteny in the studies showing benefit over harm with linially signifiant morbidity and mortality redutions with statin use that have not yet been attributed to onfounding effets. Third, urrently reommended therapies for COPD merely ontrol symptoms and have not been shown to redue morbidity or mortality onviningly. 97 Novel treatments suh as statins, that have been shown to attenuate neutrophili inflammation by reduing reruitment and/or ativation of inflammatory ells in the lungs, 97 should be onsidered. To date, approximately 20 30% of patients with COPD are presribed statins. However, given the impressive safety data on long term statin use, 91 in partiular that lowering holesterol does not inrease risk of aner nor does it inrease mortality, should we be onsidering statins more often in patients with COPD? Based on the evidene to date showing linially relevant redutions in ardiovasular mortality in high risk populations (20 30%) and the benefits of adding statins to existing COPD treatment (onsistent morbidity and mortality redutions of 30 50%), the answer is quite possibly yes. MULTIPLE CHOICE QUESTIONS (TRUE (T)/FALSE (F); ANSWERS AFTER THE REFERENCES) 1. Based on the observational studies, whih one of the following potential benefits of statin use in patients with COPD is false? A. Redued risk of lung aner B. Redued deline in FEV 1 C. Redued ardiovasular and all ause mortality Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. Postgrad Med J 2009;85: doi: /pgmj

7 D. Redued hospitalisation and mortality from infetive exaerbations E. Redution in symptoms and quality of life 2. Based on the animal and human studies, whih one of the following has not been attributed to statin effets in the lungs? A. Redued apoptosis in the lung B. Redued neutrophil influx and/or migration in to the lung C. Bronhodilatation and smooth musle relaxation D. Redued ytokine prodution in the lung E. Redued small airway inflammation 3. Whih of the following statements regarding reported observations between CRP and COPD is false? A. Elevated CRP is inversely related to impaired FEV 1 in patients with COPD, after adjustment for smoking B. Elevated CRP in COPD patients is unrelated to ardiovasular mortality C. Elevated CRP and redued FEV 1 have been assoiated in ross-setional studies D. Statin use is assoiated with a redution in CRP E. Elevated CRP is assoiation with inreased mortality in patients with COPD 4. Whih of the following statements about the use of statins in COPD is false? A. Benefits of statin use from observational studies suggest signifiant redution in respiratory morbidity even after adjustment for onventional treatment suh as ortiosteroids B. Given the high ardiovasular risk assoiated with COPD, statins may be benefiial in many patients with COPD C. Benefits from statins over existing treatments for COPD may be attributed to their apparent anti-neutrophili effets D. From observational data, it appears statins are taken by the majority of patients with COPD E. In a sub-analysis of a large randomised ontrolled trial examining statin use and their effet on ardiovasular disease, a trend for benefit on respiratory morbidity/ mortality was seen despite the inlusion of non-smokers and subjets with normal lung funtion 5. Whih of the following statements with respet to potential statin benefits in COPD is unlikely to aount for the healthy user effet? A. COPD patients taking statins have a higher prevalene of ardiovasular disease and higher mortality risk. B. COPD patients taking statins have greater impairment in lung funtion ompared with those not taking statins C. COPD patients taking statins have morbidity/mortality benefits irrespetive of onomitant use of inhaled ortiosteroids. D. COPD patients taking statins with low baseline ardiovasular risk had redutions in respiratory morbidity E. COPD patients taking statins undertake preventive interventions at a slightly higher rate than those not taking statins but is unlikely to aount for the apparent benefiial effets of statins Competing interests: None. REFERENCES 1. Vilkman S, Keistinen T, Tuuponen T, et al. Survival and ause of death among elderly hroni obstrutive pulmonary disease patients after first admission to hospital. Respiration 1997;64: Anthonisen NR, Connett JE, Enright PL, et al, Lung Health Study Researh Group. Hospitalizations and mortality in the Lung Health Study. Am J Respir Crit Care Med 2002;166: Shunemann HJ, Dorn J, Grant BJB, et al. Pulmonary funtion is a long term preditor of mortality in the general population: 29 year follow-up of the Buffalo Health study. Chest 2000;118: Doll R, Peto R, Wheatley K, et al. Mortality in relation to smoking: 40 years observations on male British dotors. BMJ 1994;309: MClay JD, MAllister DA, MaNee W. Cardiovasular risk in hroni obstrutive pulmonary disease. Respirology 2007;12: MGarvey LP, John M, Anderson JA, et al. Asertainment of ause-speifi mortality in COPD: operations of the TORCH linial endpoint ommittee. Thorax 2007;62: Hole DJ, Watt GCM, Davey-Smith G, et al. Impaired lung funtion and mortality risk in men and women: findings from the Renfrew and Paisley prospetive population study. BMJ 1996;313: Mannino DM, Buist AS, Petty TL, et al. Lung funtion and mortality in the United States: data from the First National and Nutrition Examination Survey follow up study. Thorax 2003;58: Nishimura K, Tsukino M. Clinial ourse and prognosis of patients with hroni obstrutive pulmonary disease. Curr Opinion Pulm Med 2000;6: Anthonisen NR, Connett JE, Kiley JP, et al. Effets of smoking intervention and use of inhaled antiholinergi bronhodilator on the rate of deline of FEV 1. JAMA 1994;272: Celli BR, Thomas NE, Anderson JA, et al. Effets of pharmootherapy on rate of deline of lung funtion in hroni obstrutive pulmonary disease. Am J Respir Crit Care Med 2008;178: Pauwels RA, Rabe KF. Burden and linial features of hroni obstrutive pulmonary disease (COPD). Lanet 2004;364: Vestbo J, Soorenson T, Lange P, et al. Long-term effet of inhaled budesonide in mild and moderate hroni obstrutive pulmonary disease: a randomized ontrol trial. Lanet 1999;353: Albers M, Shermer T, van den Boom G, et al. Effiay of inhaled steroids in undiagnosed subjets at high risk of COPD: results of the detetion, intervention and monitoring of COPD and asthma program. Chest 2004;126: Calverley PM, Anderson JAS, Celli B, et al. Salmeterol and flutiasone proprionate and survival in hroni obstrutive pulmonary disease. N Eng J Med 2007;356: Mannino DM, Watt G, Hole D, et al. The natural history of hroni obstrutive pulmonary disease. Eur Respir J 2006;27: Stav D, Raz M. Prevalene of hroni obstrutive pulmonary disease among smokers aged 45 and up in Israel. Isr Med Asso J 2007;9: Sin DD, Wu LL, Man SFP. The relationship between redued lung funtion and ardiovasular mortality: a population-based study and a systemati review of the literature. Chest 2005;127: Knot UM, Khot MB, Bajzer CT, et al. Prevalene of onventional risk fators in patients with oronary artery disease. JAMA 2003;290: Stavem K, Aaser E, Sandvik L, et al. Lung funtion, smoking and mortality in a 26- year follow-up of healthy middle-aged males. Eur Resp J 2005;25: Tokman MS, Comstok GW. Respiratory risk fators and mortality: longitudinal studies in Washington County, Maryland. Am Rev Respir Dis 1989;140:S Olofson J, Skoogh BE, Bake B, et al. Mortality related to smoking habits, respiratory symptoms and lung funtion. Eur J Respir Dis 1987;71: Sunyer J, Ulrik CS. Level of FEV 1 as a preditor of all-ause and ardiovasular mortality: an effet beyond smoking and physial fitness? Eur Respir J 2005;25: Flether C, Peto R. The natural history of hroni airflow obstrution. BMJ 1977;1: Lindberg A, Bjerg-baklund A, Ronmark E, et al. Prevalene and under-diagnosis of COPD by disease severity and the attributable fration of smoking. Resp Med 2006;100: Mannino DM, Pharm DT, Swensen A, et al. Prevalene and outomes of diabetes, hypertension, and ardiovasular disease in hroni obstrutive pulmonary disease. Eur Respir J 2008;32: Johnston AK, Mannino DM, Hagan GW, et al. Relationship between lung funtion impairment and inidene or reurrene of ardiovasular events in a middle-aged ohort. Thorax 2008;63: Agusti A. Systemi effets of hroni obstrutive pulmonary disease: what we know and what we don t know (but should). Pro Am Thora So 2007;4: Mattson ME, Pollak ES, Cullen JW. What are the odds that smoking will kill you? Am J Pub Health 1987;77: Lokke A, Lange P, Sharling H, et al. Developing COPD: a 25 year follow up study of the general population. Thorax 2006;61: Skillrud DM, Offard KP, Miller RD. Higher risk of lung aner in hroni obstrutive pulmonary disease: a prospetive mathed ontrolled study. Ann Intern Med 1986;105: Mannino DM, Aguayo SM, Petty TL, et al. Low lung funtion and inident lung aner in the United States: data From the First National Health and Nutrition Examination Survey follow-up. Arh Int Med 2003;163: Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. 420 Postgrad Med J 2009;85: doi: /pgmj

8 33. Islam SS, Shottenfeld D. Delining FEV 1 and hroni produtive ough in igarette smokers: a 25 year prospetive study of lung aner inidene in Teumseh, Mihigan. Caner Epidemiol Biomarkers Prev 1994;3: Loganathan RS, Stover DE, Shi W, et al. Prevalene of COPD in women ompared to men around the time of diagnosis of primary lung aner. Chest 2006;129: Wilson DO, Weissfeld JL, Balkan A, et al. Assoiation of radiographi emphysema and airflow limitation with lung aner. Am Respir Crit Care Med 2008;178: Petty TL. Are COPD and lung aner two manifestations of the same disease? Chest 2005;128: Brody JS, Spira A. Chroni obstrutive lung disease, inflammation and lung aner. Pro Am Thora So 2006;3: Shwartz AG, Rukdeshel JC. Familial lung aner: geneti suseptibility and relationship to hroni obstrutive pulmonary disease. Am J Respir Crit Care Med 2006;173: Young RP, Hopkins RJ, Eaton TE. Fored expiratory volume in one seond: not just a lung funtion test but a marker of premature death from all auses. Eur Resp J 2007;30: Amos CI, Wu X, Broderik P, et al. Genome-wide assoiation san of tag SNPs identifies a suseptibility lous for lung aner at 15q25.1. Nature Genetis 2008;40: Pillai SG, Ge D, Zhu G, et al. Genome-wide assoiation study in hroni obstrutive pulmonary disease (COPD): identifiation of two major suseptibility loi. Plos Genetis 2009;5:issue 3 online. 42. Young RP, Hopkins RJ, Hay BA, et al. Lung aner gene assoiated with COPD: triple whammy or possible onfounding effet? Eur Respir J 2008;32: Sin DD, Man PSF. Why are patients with hroni obstrutive pulmonary disease at inreased risk of ardiovasular disease? The potential role of systemi inflammation in hroni obstrutive pulmonary disease. Cirulation 2003;107: Gan WQ, Man SFP, Senthilselvan A, et al. Assoiation between hroni obstrutive pulmonary disease and systemi inflammation: a systemati review and a metaanalysis. Thorax 2004;59: Rodriguz-Roisin R, Soriano JB. Chroni obstrutive pulmonary disease with lung aner and/or ardiovasular disease. Pro Am Thora So 2008;5: Rennard SI. Inflammation in COPD: a link to systemi omorbidities. Eur Respir Rev 2007;16: Fabbri LM, Luppi F, Beghe B, et al. Complex hroni omorbidities of COPD. Eur Respir J 2008;31: Aronston D, Roterman I, Yigla M, et al. Inverse assoiation between pulmonary funtion and C-reative protein in apparently healthy subjets. Am J Respir Crit Care Med 2006;174: Shaaban R, Kony S, Driss F, et al. Change in C-reative protein levels and FEV 1 deline: a longitudinal population-based study. Resp Med 2006;100: Fogarty AW, Jones S, Britton JR, et al. Systemi inflammation and deline in lung funtion in a general population: a prospetive study. Thorax 2007;62: Dahl M, Vestbo J, Lange P, et al. C-reative protein as a preditor of prognosis in hroni obstrutive pulmonary disease. Am J Respir Crit Care Med 2007;175: de Torres JP, Pinto-Plata V, Casanova C, et al. C-reative protein levels and survival in patients with moderate to very severe COPD. Chest 2008;133: Watson L, Vonk JM, Lofdahl CG, et al. Preditors of lung funtion and its deline in mild to moderate COPD in assoiation with gender: results from the Eurosop study. Resp Med 2006;100: Brass DM, Hollingsworth JW, Cinque M, et al. Chroni LPS inhalation auses emphysema-like hanges in mouse lung that is assoiated with apoptosis. Am J Respir Cell Mol Biol 2008;39: Zhang H, Hang j, Wang X, et al. TNF polymorphisms modify endotoxin exposureassoiated longitudinal lung funtion deline. O Environ Med 2007;64: Yanbaeva DG, Dentener MA, Creutzberg EC, et al. Systemi inflammation in COPD: is geneti suseptibility a key fator? COPD: J Chr Obstr Pulm Dis 2006;3: Melbye H, Halvorsen DS, Hartz I, et al. Bronhial airflow limitation, smoking, body mass index, and statin use are strongly assoiated with the C-reative protein level in the elderly. The Tromso study Resp Med 2007;101: Ridker PM, Danielson E, Fonsea FAH, et al. Rosuvastatin to prevent vasular events in men and women with elevated C-reative protein. N Engl J Med 2008;359: Halpin D. Mortality in COPD: inevitable or preventable? Insights from the ardiovasular arena. COPD: J Chr Obstr Pulm Dis 2008;5: Shonbek U, Libby P. Inflammation, immunity, and HMG-CoA redutase inhibitors: statins as anti-inflammatory agents? Cirulation 2004;109:I Pignone M, Phillips C, Mulrow C. Use of lipid lowering drugs for primary prevention of oronary heart disease: meta-analysis of randomized trials. BMJ 2000;321: Manuel DG, Lim J, Tanuseputro P, et al. Revisiting Rose: strategies for reduing oronary heart disease. BMJ 2006;332: Hothersall E, MSharry C, Thomson NC. Potential therapeuti role for statins in respiratory disease. Thorax 2006;61: Lee JH, Lee DS, Kim EK, et al. Simvastatin inhibits igarette smoking-indued emphysema and pulmonary hypertension in rat lungs. Am J Respir Crit Care Med 2005;172: Morimoto K, Janssen WJ, Fessler MB, et al. Lovastatin enhanes learane of apoptoti ells (efferoytosis) with impliations for hroni obstrutive pulmonary disease. J Immunol 2006;176: Jaobson JR, Barnard JW, Grigoryev DN, et al. Simvastatin attenuates vasular leak and inflammation in murine inflammatory lung injury. Am J Physiol Lung Cell Mol Physiol 2005;288:L Cazzola M, Ciaprini C, Page CP, et al. Targeting systemi inflammation: novel therapies for treatment of hroni obstrutive pulmonary disease. Expert Opin Ther Targets 2007;11: Walsh GM. Statins as emerging treatments for asthma and hroni obstrutive pulmonary disease. Expert Rev Resp Med 2008;2: Hunninghake DB. Cardiovasular disease in hroni obstrutive pulmonary disease. Pro Am Thora So 2005;2: Manini GBJ, Etminan M, Zhang B, et al. Redution of morbidity and mortality by statins, angiotensin-onverting enzyme inhibitors, and angiotensin reeptor blokers in patients with hroni obstrutive pulmonary disease. J Am Coll Cardiol 2006;47: Soyseth V, Brekke PH, Smith P, et al. Statin use is assoiated with redued mortality in COPD. Eur Respir J 2007;29: Frost FJ, Petersen H, Tollestrup K, et al. Influenza and COPD mortality protetion as pleiotropi, dose-dependent effets of statins. Chest 2007;131: van Gestel YRBM, Hoeks SE, Sin DD, et al. Effet of statin therapy on mortality in patients with peripheral arterial disease and omparison of those with versus without assoiated hroni obstrutive pulmonary disease. Am J Cardiol 2008;102: Keddissi JI, Younis WG, Chbeir EA, et al. The use of statins and lung funtion in urrent and former smokers. Chest 2007;132: Mannino DM, Davis KJ. Lung funtion deline and outomes in an elderly population. Thorax 2006;61: Mannino DM, Reihert MM, Davis KJ. Lung funtion deline and outomes in an adult population. Amer J Respir Crit Care Med 2006;173: Burge PS, Calverley PMA, Jones PW, et al. Randomised, double blind, plaebo ontrolled study of flutiasone proprionate in patients with moderate to severe hroni obstrutive pulmonary disease: the ISOLDE trial. BMJ 2000;320: Watson L, Vonk JM, Lofdahl CG, et al. Preditors of lung funtion and its deline in mild to moderate COPD in assoiation with gender: results from the Eurosop study. Respir Med 2006;100: Alexeeff SE, Litonjua AA, Sparrow D, et al. Statin use redues deline in lung funtion: VA Normative Aging Study. Am J Respir Crit Care Med 2007;176: Sin DD, Man SFP. Pharmaotherapy for mortality redution in hroni obstrutive pulmonary disease. Pro Am Thora So 2006;3: Kiri VA, Pride NB, Soriano JB, et al. Inhaled ortiosteroids in hroni obstrutive pulmonary disease: results from two observational designs free of immortal time bias. Am J Respir Crit Care Med 2005;172: Khurana V, Bejjanki HR, Caldito G, et al. Statins redue the risk of lung aner in humans: a large ase-ontrol study of US veterans. Chest 2007;131: Farwell WR, Sranton RE, Lawler EV, et al. The assoiation between statins and aner inidene in a Veterans population. J Natl Caner Inst 2008;100: Karp I, Behlouli H, LeLorier J, et al. Statins and aner. Am J Med 2008;121: Mortensen EM, Pugh MJ, Copeland LA, et al. Impat of statins and angiotensinonverting enzyme inhibitors on mortality of subjets hospitalized with pneumonia. Eur Respir J 2008;31: Thomsen RW. The lesser known effets of statins: benefits on infetious outomes may be explained by healthy user effet. BMJ 2006;333: Woerndle RH, Maxwell DL. Statins and sepsis: good bullet, disappointing target. J Am Osteopath Asso 2008;108: Hurst JR, Hagan G, Wedziha JA. Mehanism of statin-assoiated mortality redution in COPD. Chest 2007;132: Heart Protetion Study Collaborative Group. The effets of holesterol lowering with simvastatin on ause-speifi mortality and on aner inidene in 20,536 high risk people: a randomized plaebo-ontrolled trial. BMC Mediine 2005;3: Majumdar SR, MAlister FA, Eurih DT, et al. Statins and outomes in patients admitted to hospital with ommunity aquired pneumonia: population based prospetive ohort study. BMJ 2006:333: Armitage J. The safety of statins in linial pratie. Lanet 2007;370: Brookhart MA, Patrik AR, Dormuth C, et al. Adherene to lipid-lowering therapy and the use of preventive health servies: an investigation of the healthy user effet. Am J Epidemiol 2007;166: Centers for Disease Control. Deaths attributable to smoking in the US. MMWR Morb Mortal Wkly Rep 2005;54: Roberts CGP, Guallar E, Rodriguez A. Effiay and safety of statin monotherapy in older adults: a meta-analysis. J Geront 2007;62A: Alsheikh-Ali AA, Trikalinos TA, Kent DM, et al. Statins, low-density lipoprotein holesterol, and risk of aner. J Am Coll Cardiol 2008;52: Ramsey SD, Hobbs FDR. Chroni obstrutive pulmonary disease, risk fators, and outome trials. Pro Am Thora So 2006;3: Rohe N. Where urrent pharmaologial therapies fall short in COPD: symptom ontrol is not enough. Eur Respir Rev 2007;16: Answers 1. A (T); B (T); C (T); D (T); E (F) 2. A (T); B (T); C (F); D (T); E (T) 3. A (T); B (F); C (T); D (T); E (T) 4. A (T); B (T); C (T); D (F); E (T) 5. A (T); B (F); C (T); D (T); E (T) Postgrad Med J: first published as /pgmj on 23 July Downloaded from on 19 June 2018 by guest. Proteted by opyright. Postgrad Med J 2009;85: doi: /pgmj