Priya Ranganathan

Common pitfalls in statistical analysis: Clinical versus statistical significance

In clinical research, study results, which are statistically significant are often interpreted as being clinically important. While statistical significance indicates the reliability of the study results, clinical significance reflects its impact on clinical practice. The third article in this series exploring pitfalls in statistical analysis clarifies the importance of differentiating between statistical significance and clinical significance.

Common pitfalls in statistical analysis: The perils of multiple testing.

Multiple testing refers to situations where a dataset is subjected to statistical testing multiple times - either at multiple time-points or through multiple subgroups or for multiple end-points. This amplifies the probability of a false-positive finding. In this article, we look at the consequences of multiple testing and explore various methods to deal with this issue.

Common pitfalls in statistical analysis: Absolute risk reduction, relative risk reduction, and number needed to treat.

In the previous article in this series on common pitfalls in statistical analysis, we looked at the difference between risk and odds. Risk, which refers to the probability of occurrence of an event or outcome, can be defined in absolute or relative terms. Understanding what these measures represent is essential for the accurate interpretation of study results.

Common pitfalls in statistical analysis: The use of correlation techniques

Correlation is a statistical technique which shows whether and how strongly two continuous variables are related. In this article, which is the eighth part in a series on ′Common pitfalls in Statistical Analysis′, we look at the interpretation of the correlation coefficient and examine various situations in which the use of technique of correlation may be inappropriate.

Common pitfalls in statistical analysis: "P" values, statistical significance and confidence intervals

In the second part of a series on pitfalls in statistical analysis, we look at various ways in which a statistically significant study result can be expressed. We debunk some of the myths regarding the ‘P’ value, explain the importance of ‘confidence intervals’ and clarify the importance of including both values in a paper

Paraplegia following epidural analgesia: A potentially avoidable cause?

Neurological deficit is an uncommon but catastrophic complication of epidural anesthesia. Epidural hematomas and abscesses are the most common causes of such neurological deficit. We report the case of a patient with renal cell carcinoma with lumbar vertebral metastasis who developed paraplegia after receiving thoracic epidural anesthesia for a nephrectomy. Subsequently, on histo-pathological examination of the laminectomy specimen, the patient was found to have previously undiagnosed thoracic vertebral metastases which led to a thoracic epidural hematoma. In addition, delayed reporting of symptoms of neurological deficit by the patient may have impacted his outcome. Careful pre-operative investigation, consideration to using alternative modalities of analgesia, detailed patient counseling and stringent monitoring of patients receiving central neuraxial blockade is essential to prevent such complications.

Meta-analysis: Adding apples and oranges?

Sir, I read with interest the results of the meta-analysis by Krishna et al.,[1] on the role of noninvasive positive pressure ventilation in postextubation respiratory failure and I commend the authors on their detailed and comprehensive methodology and well-written paper. However, I would like to share a couple of suggestions which are crucial to the interpretation of the results of this meta-analysis. The authors found moderate heterogeneity between included studies (I2 statistic of 48.5%). Summary statistics for a meta-analysis can be calculated using two types of statistical models: fixed-effects, when there is minimal heterogeneity and random-effects, when there is a higher level of heterogeneity. Of these, the random-effects model is more conservative and in the setting of heterogeneity, is likely to give more dependable results.[2] For better understanding and to allow readers to judge the validity of the results, the authors should have specified the type of analysis which was used in this review. The quality of any systematic review or meta-analysis is only as good as that of the included studies. The authors have done a quality assessment of the studies incorporated in this review and it appears that several of the studies were not of good quality (score of 16 or less). In particular, one study by Jiang not only had a low quality (score of 13) but also had results which were diametrically opposite to the other studies in the analysis. This study was given a high weightage of 20% in calculating the overall summary statistic. It is surprising that the authors did not conduct a sensitivity analysis by excluding this particular study, and also another analysis by separating good from poor-quality studies. This would probably have made the results of the meta-analysis more precise and reliable.

Appropriateness of perioperative blood transfusion in patients undergoing cancer surgery: A prospective single-centre study.

Background: Allogenic blood transfusion is associated with several potential complications, especially in patients with cancer. The objective of this prospective single-centre study was to identify the rates of perioperative blood transfusion and overtransfusion in a tertiary-level cancer hospital. Methods: Between March and May 2008, we studied all adult patients undergoing elective major cancer surgery under anaesthesia and recorded intra- and immediate post-operative (within 24 h) blood transfusions and post-operative investigations. Overtransfusion was defined as post-transfusion haemoglobin (Hb) exceeding 10 g/dL. Results: One hundred and eighty-six of 1175 (16%) patients received perioperative blood transfusion. The main trigger for intraoperative transfusion was blood loss exceeding the patients maximum allowable blood loss (92, 49%). Ninety-five (51%) transfused patients had post-transfusion Hb more than 10 g/dL. The rate of overtransfusion was not higher in patients who received single-unit transfusions. Conclusion: The perioperative transfusion rate in patients undergoing cancer surgery was 16%. More than half of these patients were overtransfused. Following this audit, point-of-care facilities for intraoperative haemoglobin measurement have been introduced.

Bronchial blocker for one-lung ventilation: An unanticipated complication

Techniques described for achieving lung isolation for thoracic surgery include the use of double-lumen tubes (DLT) and bronchial blockers. The bronchial blocker has several advantages over DLTs, especially in the setting of the difficult airway.[1] However, potential disadvantages include longer time for placement, longer time for lung collapse and poor quality of suctioning through the blocker.[2]

Pain after posterolateral versus nerve-sparing thoracotomy: A randomized trial

Objectives: Post‐thoracotomy pain leads to patient discomfort, pulmonary complications, and increased analgesic use. Intercostal nerve injury during thoracotomy or its entrapment during closure can contribute to post‐thoracotomy pain. We hypothesized that a modified technique of posterolateral thoracotomy and closure, preserving the intercostal neurovascular bundle, would reduce acute and chronic post‐thoracotomy pain. Methods: We randomized 90 patients undergoing posterolateral thoracotomy for pulmonary resection at a tertiary level oncology center to standard posterolateral (control arm) or modified nerve‐sparing thoracotomy. All patients received morphine via patient‐controlled analgesia pumps. The primary outcome was the worst postoperative pain score in the first 3 postoperative days. Secondary outcomes included the average pain score and analgesic requirements in the first 3 postoperative days and the incidence of post‐thoracotomy pain 6 months after surgery. Results: No significant differences were seen between the groups in acute or chronic post‐thoracotomy measured by the numeric rating scale. There was no difference seen in the worst (mean) postoperative pain scores (3.71 vs 3.83, difference 0.12; 99% confidence interval [CI], −0.7 to +0.9; P = .7), average (mean) pain scores in the first 3 postoperative days (1.77 vs 1.85, difference 0.08; 99% CI, −0.4 to +0.6; P = .69), mean consumption of morphine (mg/kg) (1.45 vs 1.40, difference −0.05; 99% CI, −0.4 to +0.3; P = .73), or incidence of chronic postoperative pain (37.8% vs 40%, difference 4.9%; 99% CI, −22.8 to +30.7%; P = .73). Conclusions: The modified nerve‐sparing thoracotomy technique does not reduce post‐thoracotomy pain compared with standard posterolateral thoracotomy.