|LETTER TO EDITOR
|Year : 2007 | Volume
| Issue : 2 | Page : 159
Endophthalmitis research group study: Selection bias skews results
Himanshu Deshmukh, Bhagyashri Deshmukh, Pukhraj Rishi
Medical Research Foundation, Sankara Nethralaya, Chennai, India
Medical Research Foundation, Sankara Nethralaya, Chennai
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Deshmukh H, Deshmukh B, Rishi P. Endophthalmitis research group study: Selection bias skews results. Indian J Ophthalmol 2007;55:159
|How to cite this URL:|
Deshmukh H, Deshmukh B, Rishi P. Endophthalmitis research group study: Selection bias skews results. Indian J Ophthalmol [serial online] 2007 [cited 2019 Aug 23];55:159. Available from: http://www.ijo.in/text.asp?2007/55/2/159/30719
We read the article by Das et al . with interest. Their study provides valuable information and fills previous gaps about endophthalmitis in India.
The authors based their assessment of the relationship between the clinical presentation and visual outcome in postoperative endophthalmitis on a prospective analysis of 128 cases. All patients in the study received vitrectomy. The authors are not clear whether any patients were managed by intravitreal antibiotics alone. Neither do they mention conservative management of endophthalmitis as their exclusion criteria. This leads us to make the following observations,
- The authors did not include patients who were managed by intravitreal antibiotics alone. We know that, according to endophthalmitis vitrectomy study (EVS) guidelines, patients with good vision and less severe endophthalmitis are managed with vitreous tap and intravitreal antibiotics alone and do not need a vitrectomy. And obviously patients with good vision and less severe endophthalmitis at presentation have a better final visual outcome. By excluding these patients from the study and incorporating a selection bias, the results would be skewed towards a worse visual outcome. Any study which intends to elucidate the relationship between the clinical presentation and the final outcome has to include the entire spectrum of the disease and not exclude those cases in which vitrectomy was not performed on account of patients not needing it.
- Even among the 128 patients analyzed in the postoperative endophthalmitis group, 9.4% patients had a visual acuity of >20/400 and another 30.5% had a visual acuity of hand movements (HM). The EVS suggests patients with visual acuity ³ HM can be managed conservatively with intravitreal antibiotics alone, without resorting to vitreous surgery. We would like to know the reasons behind subjecting these patients to a surgical procedure.
As the authors' article is the first, large prospective study of endophthalmitis conducted in India, it is important that their message is clearly passed on to the target audience. An explanation in this regard would greatly clarify things regarding postoperative endophthalmitis in India.
| References|| |
Das T, Kunimoto DY, Sharma S, Jalali S, Majji AB, Nagaraja Rao T, et al
. Endophthalmitis Research Group. Relationship between clinical presentation and visual outcome in postoperative and posttraumatic endophthalmitis in south central India. Indian J Ophthalmol
Results of the Endophthalmitis Vitrectomy Study. A randomized trial of immediate vitrectomy and of intravenous antibiotics for the treatment of postoperative bacterial endophthalmitis. Endophthalmitis Vitrectomy Study Group. Arch Ophthalmol