Intraoperative floppy iris syndrome (IFIS) is associated with the intake of α1-adrenergic receptor antagonists (α1-blockers) (Chang et al. 2008). Our recent meta-analysis has highlighted a hierarchy in the magnitude of the associations between IFIS and various α1-blockers (Chatziralli & Sergentanis 2010).

This Letter aims to evaluate risk factors for IFIS in patients undergoing phacoemulsification. All patients (n = 738, Caucasian) who underwent phacoemulsification cataract surgery from 1 January 2008 to 23 January 2010 at Veroia General Hospital were included. The following data were evaluated as possible risk factors: ophthalmological conditions, sociodemographic features, eye colour (routinely collected in the medical records), clinical data (hypertension, diabetes mellitus, renal failure, heart failure) and medications being taken at the time of surgery. The information regarding medication was gathered by the medical health records of patients (physicians’ archives and insurance data). To consider “current use” of a medication as meaningful, at least one month of intake should have been documented. In addition, concerning especially α1-blockers, ever-use of the agents was recorded, so as to perform an alternative analysis taking into account any carry-over effects. Cases were characterized intraoperatively as IFIS (complete or incomplete, according to Chang et al. 2008) and non-IFIS. Univariate and multivariate logistic regression analyses were performed. Statistical analysis was performed with STATA version 9.1 (STATA Corp., College Station, TX, USA). This study is in accordance with the Declaration of Helsinki and has been approved by the local Institutional Review Board.

IFIS was observed in 43 of 738 eyes (5.8%). Among 43 cases, 22 pertained to complete IFIS and 21 to incomplete IFIS. At the univariate analysis, male sex, heart failure, current α1-blocker use (tamsulosin, alfuzosin, terazosin), current benzodiazepine use and finasteride use were associated with IFIS. Current use of rivastigmine and of quetiapine was associated with IFIS at a borderline level. However, no significant association implicated age, eye colour, diabetes mellitus, hypertension and pseudoexfoliation.

Table 1 presents the results of the multivariate analysis. Current use of tamsulosin, alfuzosin, terazosin, benzodiazepines and finasteride were independently associated with IFIS. Noticeably, male sex and heart failure lost their significance at the multivariate approach. Concerning ever-use of α1-blockers, the results of the alternative analysis remained practically similar to the analysis on current use.

Table 1.   Variables retained in the multivariate logistic regression analysis (backward selection), with their respective odds ratios (mutually adjusted).
VariableCategory or incrementOdds ratio (95% confidence interval)p
Current tamsulosin useYes versus no4058.0 (344.1–47850.2)<0.001
Current alfuzosin useYes versus no394.5 (67.5–2304.8)<0.001
Current terazosin useYes versus no507.7 (52.8–4880.9)<0.001
Current benzodiazepine useYes versus no676.5 (133.0–3440.9)<0.001
Current finasteride useYes versus no338.2 (30.6–3733.8)<0.001

This retrospective study is the first to demonstrate an independent, positive association between benzodiazepines and IFIS. A biochemical link may well exist, as benzodiazepine receptors are present in various regions of the eye including the ciliary body and specifically the iris (Zarbin & Anholt 1991). Noticeably, given that phacoemulsification may be accompanied by considerable anxiety, this finding may entail an aspect with clinical significance, i.e. anxious patients should be discouraged against receiving benzodiazepines during the days prior to phacoemulsification.

Similarly to previous case reports on finasteride (Issa & Dagres 2007), none of our four patients reported the current use of α1-blockers; consequently, the effect of finasteride seems independent. Secondary, the borderline association between IFIS and use of antipsychotic drugs is in agreement with case reports (Papadopoulos & Bachariou 2007); impressively enough, the mechanism of action of quetiapine entails antagonism at α1-adrenergic receptors (Tanibuchi et al. 2009). The finding implicating rivastigmine parallels a case report on another closely related acetylcholinesterase inhibitor, donepezil (Papadopoulos & Bachariou 2007).

In conclusion, this study confirms the hierarchy of α1-blockers and additionally points to current benzodiazepine use as a risk factor for IFIS. Caution is also advised for patients receiving finasteride, rivastigmine or donepezil. Future studies should evaluate ever-use of benzodiazepines apart from current use.


  1. Top of page
  2. References