Important Methodological Flaws Research Articles

Laparoscopic Versus Open Appendectomy in Adults with Complicated Appendicitis: Systematic Review and Meta‐analysis

The goal of the present study was to critically review and identify the strength of available evidence in the literature on the use of laparoscopic appendectomy (LA) in complicated appendicitis (CA). The Cochrane Library and Controlled Trials Registry, MEDLINE (Ovid), PubMed, Web of knowledge, and SCOPUS databases were electronically searched, using the keywords "appendectomy," "laparoscopy," "appendicitis." "complicated appendicitis." "gangrenous appendicitis," "perforated appendicitis," with English language as a limit. Backward chaining was also employed. The NHS Public Health Resource Unit Critical Appraisal Skills Programme Tools were used for critical appraisal. Twelve retrospective case-control studies were included in the review. Overall methodological quality was moderate to poor, with heterogeneity, absence of randomization and blinding, and presence of important methodological flaws. Meta-analysis showed that LA in CA has reduced surgical site infection (SSI) rates compared to open appendectomy (OA), odds ratio (OR) 0.23, 95% confidence intervals (CI): 0.14-0.37 (level 3a evidence), and no difference with regard to intra-abdominal abscess (IAA) complication rates OR: 1.02, 95% CI 0.56-1.86 (level 3a evidence). When compared to OA, laparoscopic appendectomy is advantageous in CA with regard to SSIs, with no significant additional risk of IAA (level 3a evidence).

Randomized clinical trials in implant therapy: relationships among methodological, statistical, clinical, paratextual features and number of citations

The aim of the present study was to investigate the relationships among reported methodological, statistical, clinical and paratextual variables of randomized clinical trials (RCTs) in implant therapy, and their influence on subsequent research. The material consisted of the RCTs in implant therapy published through the end of the year 2000. Methodological, statistical, clinical and paratextual features of the articles were assessed and recorded. The perceived clinical relevance was subjectively evaluated by an experienced clinician on anonymous abstracts. The impact on research was measured by the number of citations found in the Science Citation Index. A new statistical technique (Structural learning of Bayesian Networks) was used to assess the relationships among the considered variables. Descriptive statistics revealed that the reported methodology and statistics of RCTs in implant therapy were defective. Follow-up of the studies was generally short. The perceived clinical relevance appeared to be associated with the objectives of the studies and with the number of published images in the original articles. The impact on research was related to the nationality of the involved institutions and to the number of published images. RCTs in implant therapy (until 2000) show important methodological and statistical flaws and may not be appropriate for guiding clinicians in their practice. The methodological and statistical quality of the studies did not appear to affect their impact on practice and research. Bayesian Networks suggest new and unexpected relationships among the methodological, statistical, clinical and paratextual features of RCTs.

Corticosteroids for myasthenia gravis.

Although widely accepted as an appropriate immunosuppressive therapy, the efficacy of glucocorticosteroid treatment has only rarely been tested in controlled studies. To assess the efficacy of glucocorticosteroids or adrenocorticotrophic hormone (ACTH) medication in autoimmune myasthenia gravis. We searched the Cochrane Neuromuscular Disease Group Trials Register in July 2004, MEDLINE (from January 1966 to June 2004) and EMBASE (from January 1980 to June 2004). We also checked the bibliographies in reviews and the randomised trials and contacted their authors to identify additional published and unpublished data. From the articles identified we selected those open or controlled studies which allowed us to assess the outcome of treated and untreated patients at definite endpoints. Types of studies: quasi-randomised or randomised controlled trials. patients with myasthenia gravis of all ages and all degrees of severity. Types of interventions: any form of glucocorticosteroids or adrenocorticotrophic hormone treatment. Types of outcome measures:Primary outcome(1) improvement after at least three months in either the weakest muscles or all muscles. Secondary outcomes(1) proportion of patients improved after at least six months(2) proportion of patients in remission(3) number of episodes of worsening during the first six months(4) acetylcholine receptor antibody titres after at least three months of therapy. Three authors extracted the data from the selected articles and one other checked them. A trial of adrenocorticotrophic hormone (43 patients) did not show any advantage compared with placebo for the treatment of ocular myasthenia gravis. Two double-blind trials compared prednisone with placebo for generalised myasthenia gravis. In the first (13 patients), the improvement was slightly greater in the prednisone group at six months. In the second (20 patients) which was a short-term trial, the improvement was significantly greater at two weeks. Two trials compared glucocorticosteroids with azathioprine (41 and 10 patients respectively). In one of these the rate of treatment failure was greater in the prednisone group. In a trial of glucocorticosteroids versus intravenous immunoglobulin (33 patients) no differences in treatment responses were encountered during a treatment period of 14 days. An open trial (39 patients) evaluating different corticosteroid doses revealed a shorter time to improvement in the high-dose group. However only limited evidence can be drawn from the available randomised controlled trials due to numerous and important methodological flaws. Limited evidence from randomised controlled trials suggests that corticosteroid treatment offers significant short-term benefit in myasthenia gravis compared with placebo. This supports the conclusions of observational studies and expert opinion. Limited evidence from randomised controlled trials does not show any difference in efficacy between corticosteroids and either azathioprine or intravenous immunoglobulin.

Zuclopenthixol acetate for acute schizophrenia and similar serious mental illnesses.

Medication used for acute aggression in psychiatry must have rapid onset of effect, low frequency of administration and low levels of adverse effects. Zuclopenthixol acetate is said to have these properties. To estimate the clinical effects of zuclopenthixol acetate for the management of acute aggression or violence thought to be due to serious mental illnesses, in comparison to other drugs used to treat similar conditions. We supplemented past searches of Current Controlled Trials (10/2000), the Cochrane Library (1997) and MEDLINE (1966-1997) and appeals for unpublished data with an update search of the Cochrane Schizophrenia Group's Register of trials (September 2003). All randomised clinical trials involving people thought to have serious mental illnesses comparing zuclopenthixol acetate with other drugs. Data were extracted independently by two reviewers and cross-checked. We calculated fixed effects relative risks (RR) and 95% confidence intervals (CI) for dichotomous data. Where possible, the number needed to treat/harm statistic (NNT/H) was calculated. We analyzed by intention-to-treat. Mean differences were used for continuous variables. We found no data for the primary outcome, tranquilisation. Compared with haloperidol, zuclopenthixol acetate was no more sedating at two hours (n=40, 1 RCT, RR 0.60 CI 0.27 to 1.34). People given zuclopenthixol acetate were not at reduced risk of being given supplementary antipsychotics (n=134, 3 RCTs, RR 1.49 CI 0.97 to 2.30) although additional use of benzodiazepines was less (n=50, 1 RCT, RR 0.03 CI 0.00 to 0.47, NNT 2 CI 2 to 4). People given zuclopenthixol acetate had fewer injections over seven days compared with those allocated to haloperidol IM (n=70, 1 RCT, RR 0.39 CI 0.18 to 0.84, NNT 4 CI 3 to 14). We found no data on more episodes of aggression or harm to self or others. One trial (n=148) reported no significant difference in adverse effects for people receiving zuclopenthixol acetate compared with those allocated haloperidol at one, three and six days (RR 0.74 CI 0.43 to 1.27). Compared with haloperidol or clotiapine, people allocated zuclopenthixol did not seem to be at more risk of a range of movement disorders (<20%). Three studies found no difference in the proportion of people getting blurred vision/ dry mouth (n=192, 2 RCTs, RR at 24 hours 0.90 CI 0.48 to 1.70). Similarly dizziness was equally infrequent for those allocated zuclopenthixol acetate compared with haloperidol (n=192, 2 RCTs, RR at 24 hours 1.15 CI 0.46 to 2.88). There was no difference between treatments for leaving the study before completion (n=522, RR 0.85 CI 0.31 to 2.31). Recommendations on the use of zuclopenthixol acetate for the management of psychiatric emergencies in preference to 'standard' treatment have to be viewed with caution. Most trials present important methodological flaws and findings are poorly reported. This review did not find any suggestion that zuclopenthixol acetate is more or less effective in controlling aggressive acute psychosis, or in preventing adverse effects than intramuscular haloperidol, and neither seemed to have a rapid onset of action. Well-conducted pragmatic randomised controlled trials are needed.

Zuclopenthixol acetate in the treatment of acute schizophrenia and similar serious mental illnesses.

People with schizophrenia or other psychotic illnesses may have delusions or hallucinations that may lead them to be aggressive or violent to themselves or others. Medication that is used in this context requires the properties of rapid onset of effect (tranquillisation or at least initial sedation in order to quell aggressive or disorganised behaviour, but also antipsychotic effect), low frequency of administration and low levels of side effects, such as cardiological or neurological side effects, or pain at the injection site. Zuclopenthixol is the cis(Z)-isomer of clopenthixol, a neuroleptic of the thioxanthene group, used for treating people with psychotic symptoms. There is one oral preparation and two depot forms: zuclopenthixol acetate and zuclopenthixol decanoate. The acetate version does not stay in the body for very long (a single dose persists for only 72 hours) and is said to have these properties. To estimate the effectiveness of zuclopenthixol acetate for the acute treatment of serious mental illnesses in comparison to other neuroleptic drugs. Searches of Current Controlled Trials (http://www.controlled-trials.com - accessed 5.10.2000), Cochrane Schizophrenia Group's Register of Trials (January 2001), the Cochrane Library (1997, CD-ROM, issue 2), MEDLINE (1966-1997) were supplemented by appeals for unpublished data to the research community and to the Medical Information Department of Lundbeck Limited. Attempts were made to contact relevant authors. Two reviewers independently assessed citations or papers and selected all randomised trials that included people with serious mental illnesses and compared zuclopenthixol acetate with other drug regimes. Two reviewers extracted data independently. Attempts were made to contact authors for additional or missing information. Odds ratios (OR) and 95% confidence intervals (CI) were estimated for binary data. Where possible, OR were pooled using Peto method and intention-to-treat analysis undertaken. Mean differences were used for continuous variables. Pooled data show no difference for the outcome 'no important improvement' in psychotic symptoms at the end of the follow-up period (OR 0.84 CI 0.34-2.05). Sedation was evaluated using different instruments. Only one study presented data which suggested an earlier and more intense sedation in zuclopenthixol acetate users at four hours (OR 0.18 CI 0.04-0.93). Use of additional antipsychotic medication was not avoided in the zuclopenthixol acetate group (OR 2.18, CI 0.64-7.42) and data on total number of administrations were not obtainable. Side effect data were poorly reported but there is no evidence of a consistent difference between zuclopenthixol acetate and other 'standard drugs' for either the pattern of side effects or the wish to leave the study early. Hospital and service outcomes, number of aggressive incidents, satisfaction with care and economic outcomes were not addressed by any study. Recommendations on the use of zuclopenthixol acetate for the management of psychiatric emergencies in preference to 'standard' treatment have to be viewed with caution. Most trials present important methodological flaws and findings are poorly reported. This review did not find any suggestion that zuclopenthixol acetate is more effective in controlling aggressive/disorganised behaviour, acute psychotic symptoms, or preventing side effects. There were no data directly related to tranquillisation, but it may produce more earlier and intense sedation than oral haloperidol. Well-conducted randomised controlled trials are needed to confirm claims related to the use of zuclopenthixol acetate in emergency psychiatry.

Clotiapine for acute psychotic illnesses.

Acute psychotic illness, especially when associated with agitated or violent behaviour, requires urgent pharmacological tranquillisation or sedation. Clotiapine, a dibenzothiazepine neuroleptic, is being used for this purpose in several countries. To estimate the effects of clotiapine when compared to other 'standard' or 'non-standard' treatments of acute psychotic illness in controlling disturbed behaviour and reducing psychotic symptoms. The Cochrane Controlled Trials Register (Issue 2, 2000), The Cochrane Schizophrenia Group's Register (May 2000), EMBASE (1980-2000), MEDLINE (1966-2000), PASCAL (1973-2000) and PsycLIT (1970-2000) were methodically searched. This was supplemented by hand searching reference lists, contacting industry and relevant authors. Randomised clinical trials comparing clotiapine to any treatment, for people with acute psychotic illnesses such as in schizophrenia, schizoaffective disorder, mixed affective disorders, manic phase of bipolar disorder, brief psychotic episode or organic psychosis following substance abuse. Studies were reliably selected, quality assessed and data extracted. Data were excluded where more than 50% of participants in any group were lost to follow up. For binary outcomes a standard estimation of the risk ratio (RR) and its 95% confidence interval (CI) was calculated. Where possible, the weighted number needed to treat statistic (NNT), and its 95% confidence interval (CI), was also calculated. If heterogeneity was found, a random effects model was used. For continuous outcomes, endpoint data were preferred to change data. Non-skewed data from valid scales were summated using a weighted mean difference (WMD). Again, if heterogeneity was found a random effects model was used. A Mantel-Haenszel chi-square test was used to investigate the possibility of heterogeneity. Five trials were included. None compared clotiapine with placebo, but control drugs were either antipsychotics (chlorpromazine, perphenazine, trifluoperazine and zuclopenthixol acetate) or benzodiazepines (lorazepam). Versus antipsychotics: results for global clinical outcome were heterogeneous (p=0.09) but did not suggest clotiapine to be superior, or inferior, to chlorpromazine, perphenazine, or trifluoperazine (total randomised = 83). Use of clotiapine did change the proportion of people ready for hospital discharge by the end of the study in one small trial (n=49, RR 1.04 95%CI 0.96 to 2.12). Overall, attrition rates were low. No significant difference was found for those allocated to clotiapine compared with people randomised to other antipsychotics (n=121, RR 2.26 95%CI 0.40 to 13). Weak data suggests that clotiapine may result in less need for antiparkinsonian treatment compared with zuclopenthixol acetate (n=38, RR 0.43 95%CI 0.02 to 0.98). Versus lorazepam: when used to control aggressive/violent outbursts for people already treated with haloperidol, clotiapine did not significantly improve mental state compared to lorazepam (WMD -3.36 95%CI -8.09 to 1.37). Much data could not be pooled due to skew or inadequate presentation of results. Economic outcomes and satisfaction with care were not addressed. We found no significant evidence to support the use of clotiapine rather than other 'standard' or 'non-standard' treatments for the management of acute psychotic illness. The trials included in this review all present important methodological flaws. We do not wish to discourage clinicians from using clotiapine in the psychiatric emergency, we would just like to point out the fact that good quality controlled trials are needed on this subject.

Brief or minimal intervention for ‘alcoholics’? The evidence suggests otherwise

The evidence on the relative impact of intensive intervention(s) and brief advice for alcoholic populations has been interpreted as indicating that there is little advantage for the intensive form of treatment over brief advice. The current review suggests that this conclusion is based on studies with serious and important methodological flaws, and that a slightly different, qualified interpretation is appropriate. There is evidence that intensive intervention does benefit alcohol-dependent patients, although it seems likely that the less alcohol-dependent (problem or excessive) drinkers do not require such intense intervention(s). There is also reasonable evidence that at the end of intervening with people with alcohol problems there should be some continued assistance (ie 'aftercare') available to extend the intervention, and this assistance should be structured and scheduled.

Preoperative Spirometry Before Abdominal Operations, A Critical Appraisal of Its Predictive Value

• Preoperative spirometry is commonly ordered before abdominal surgery, with the goal of predicting and preventing postoperative pulmonary complications. We assessed the evidence for this practice with a systematic literature search and critical appraisal of published studies. The search identified 135 clinical articles, of which 22 (16%) were actual investigations of the use and predictive value of preoperative spirometry. All 22 studies had important methodological flaws that preclude valid conclusions about the value of screening preoperative spirometry. The available evidence indicates that spirometry's predictive value is unproved. Unanswered questions involve (1) the yield of spirometry, in addition to history and physical examination, in patients with clinically apparent lung disease; (2) spirometry's yield in detecting surgically important occult disease; and (3) its utility, or beneficial effect on patient outcome. Spirometry's full potential for risk assessment in the individual patient has not yet been realized. (Arch intern Med1989;149:280-285)

Preoperative Spirometry Before Abdominal Operations

Preoperative spirometry is commonly ordered before abdominal surgery, with the goal of predicting and preventing postoperative pulmonary complications. We assessed the evidence for this practice with a systematic literature search and critical appraisal of published studies. The search identified 135 clinical articles, of which 22 (16%) were actual investigations of the use and predictive value of preoperative spirometry. All 22 studies had important methodological flaws that preclude valid conclusions about the value of screening preoperative spirometry. The available evidence indicates that spirometry's predictive value is unproved. Unanswered questions involve (1) the yield of spirometry, in addition to history and physical examination, in patients with clinically apparent lung disease; (2) spirometry's yield in detecting surgically important occult disease; and (3) its utility, or beneficial effect on patient outcome. Spirometry's full potential for risk assessment in the individual patient has not yet been realized.