Purpose
To compare the cataract surgery–related complications between patients with and without tamsulosin treatment.
Design
A nationwide retrospective case-control study.
Methods
Patients who had undergone cataract surgery were identified using the International Classification of Disease, Ninth Revision, Clinical Modification from a nationally representative dataset of 1 million people selected from the Taiwan National Health Insurance Research Database in 2000. Patients preoperatively treated with α 1 -blockers before cataract surgery were the treated group, and age-, sex-, and year of surgery–matched patients not preoperatively treated with α 1 -blockers were the control group. Patients treated with tamsulosin underwent subgroup analysis. A conditional logistic regression model was used to estimate surgery-related complications and interesting variables. The main outcome measures are cataract surgery–related complications.
Results
A total of 4474 treated patients and 4474 controls were analyzed. The percentage of cataract surgery–related complications was 8.61% in the treated group and 8% in the control group (not significantly different). However, wound dehiscence was 3.81 times higher (95% confidence interval: 1.24–11.67, P = .0194) in the tamsulosin-treated group.
Conclusions
Patients treated with tamsulosin have a higher risk of wound dehiscence after cataract surgery. Carefully taking a history of tamsulosin use before cataract surgery is advised so that some strategies can be used to prevent complications and additional costs.
Cataract is one of the most common eye disorders that cause vision loss, and cataract surgery is a general eye procedure worldwide. Age is the most common cause of cataract; others are medications like corticosteroids, autoimmune disease, smoking, exposure to sunlight, or a previous eye injury.
An α 1 -blocker is a type of antihypertensive drug that relaxes vascular smooth muscle cells. It can also relax urethral sphincters and is often prescribed for dysuria, especially in men with benign prostate hypertrophy. There are several kinds of α 1 -blockers: doxazosin, prazosin, terazosin, alfuzosin, and tamsulosin. Tamsulosin has a higher affinity for α 1 A receptors than do other α1-blockers, and it is usually prescribed for patients without hypertension because it has a small effect on blood pressure. A special complication of cataract surgery called “intraoperative floppy iris syndrome” (IFIS) has been reported in the last decade. IFIS has 3 clinical signs: (1) progressive pupil constriction during cataract surgery; (2) an iris that appears floppy as it billows during normal irrigation and aspiration in the anterior chamber of the operated eye; and (3) a tendency for the iris to prolapse into the phacoemulsification and side port incisions throughout surgery. IFIS can occur in patients who take any kind of α 1 -blocker (incidence: ca. 2%), but it is most common in patients who take tamsulosin.
Most published studies on IFIS associated with tamsulosin are reported in the United States and Europe, and only 1 is based on an Asian population. Moreover, although most studies report the incidence of IFIS, only a few have evaluated other surgery-related complications. Therefore, we studied the serious post–cataract surgery complications other than IFIS in patients treated with tamsulosin.
Methods
Taiwan’s nationwide National Health Insurance Research Database (NHIRD) was the data source for this retrospective case-control study. We estimated the risk of patients who had undergone cataract surgery with and without preoperative α 1 -blocker treatment. Taiwan’s National Health Insurance program began in 1995, and approximately 99% of the population is currently enrolled in the program. Personal information is confidential because patient identification numbers and other sensitive personal data are encrypted. An exemption was obtained from the Institutional Review Board of Chi Mei Medical Center (Applicant number: 10203-E01) because of the encryption.
Participants
The detailed data (demographics, disease diagnoses, contracted healthcare institutions, medical expenditures, and prescriptions) for the participants of this study were acquired from the Longitudinal Health Insurance Database 2000 (LHID2000), a random sample of 1 million people (4.5% of Taiwan’s population) from the registry of the NHIRD in 2000. Diagnoses were identified using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic and procedural codes.
Inpatients and outpatients who underwent a cataract operation (ICD-9-CM: 13.1–13.5) between January 1, 1997, and December 31, 2008, were selected for this study. Patients with a history of glaucoma (ICD-9-CM: 365) before the cataract operation and patients who had undergone other eye surgeries (ICD-9-CM: 13.8, 13.9, 14.0, 14.4, 14.5, 14.6, 14.7) within 3 years before cataract surgery were excluded. The medical records for these patients screened for having used the α1-blockers tamsulosin, alfuzosin, doxazosin, terazosin, and prazosin were identified by case group. For every patient who had undergone cataract surgery with a preoperative α1-blocker, 1 age-, sex-, and year of surgery (±1 year)–matched patient who had undergone cataract surgery without an α-1-blocker was selected as a control.
Identification of Complications
Cataract surgery–related complications were identified using the following 8 groups of ICD-9-CM diagnostic and procedure codes: glaucoma (ICD-9-CM: 365), endophthalmitis (ICD-9-CM: 360.0, 998.59, 998.69), vitreous hemorrhage (ICD-9-CM: 379.23), retinal detachment (ICD-9-CM: 14.3–14.5,14.9), intraocular lens dislocation (ICD-9-CM: 13.70, 13.72, 13.8), dropped nucleus (ICD-9-CM: 13.9, 14.70–14.79), pseudophakic corneal edema (ICD-9-CM: 10.44, 11.53, 11.60–11.64,11.69, 14.49), and wound dehiscence (ICD-9-CM: 11.5, 11.51, 11.52, 11.59, 12.66, 12.83). The last 5 groups of codes for the complications listed above were defined as “cataract reoperation–related complications” in our study; they were determined in a series cohort study in Western Australia. We used these complications because they often require hospitalization or surgical intervention and they were more reliable than the first 3 groups. Ocular hypertension (ICD-9-CM: 365.04) was used to analyze increased intraocular pressure without glaucomatous optic nerve damage after surgery. Inpatients and outpatients who had a diagnosis or procedure code for at least 1 of these complications or procedures within 2 weeks after cataract surgery were defined as having cataract surgery–related complications.
Risk Variables
The patient’s age at the time of the cataract operation, sex, hospital level, level of urbanization, and comorbidities—hypertension (ICD-9-CM: 401–405, 437.2, 362.11), diabetes mellitus (ICD-9-CM: 250), and autoimmune diseases possibly requiring steroid therapy (ICD-9-CM: 340, 696, 710.0, 714) within 1 year—were estimated in this study. The definition of hospital level was based on volume of medical personnel and beds, healthcare quality, medical education, etc, from hospital accreditation records of the Taiwan Joint Commission on Hospital Accreditation. Urbanization level was based on the population density (people/km 2 ), population ratio of people with college-or-above educational levels, population ratio of elderly people (≥65 years old), population ratio of agricultural workers, and the number of physicians per 100 000 people. Level 1 indicates the highest level of urbanization.
Statistical Methods
Student t test for continuous variables and Pearson χ 2 test or Fisher exact test for categorical variables were used to compare differences between cases and controls. The association between variables of interest and complications was measured using a conditional logistic regression model with a dependent outcome: α1-blockers. The odds ratios (ORs) and 95% confidence intervals (CIs) were calculated after adjustments for potential confounders had been made. Significance was set at P < .05. SAS 9.3 for Windows (SAS Institute, Inc, Cary, North Carolina, USA) was used for all statistical analyses.
Results
We identified 23 873 patients who had undergone cataract surgery. After age, sex, and year of surgery matching, we enrolled 8948 patients: 4474 had taken an α 1 -blocker and 4474 had not. Patients who had taken an α 1 -blocker lived in areas with a higher level of urbanization and had more hypertension and diabetes mellitus ( Table 1 ).
| α 1 -Blocker (n = 4474) | No α 1 -Blocker (n = 4474) | P Value | |
|---|---|---|---|
| Age (y) (mean ± SD) | 71.83 ± 7.92 | 71.83 ± 7.92 | .9979 |
| Age group (%) | |||
| ≤65 | 770 (17.21) | 766 (17.12) | .9107 |
| >65 | 3704 (82.79) | 3708 (82.88) | |
| Sex (%) | |||
| Female | 1020 (22.80) | 1020 (22.80) | 1.0000 |
| Male | 3454 (77.20) | 3454 (77.20) | |
| Medical level (%) | |||
| Medical center | 1227 (27.43) | 968 (21.64) | <.0001 |
| Regional hospital | 744 (16.63) | 656 (14.66) | |
| Local hospital | 2503 (55.95) | 2850 (63.70) | |
| Urbanization | |||
| 1 | 1213 (27.11) | 1155 (25.82) | .0218 |
| 2 | 2167 (48.44) | 2097 (46.87) | |
| 3 | 411 (9.19) | 458 (10.24) | |
| 4 | 683 (15.27) | 764 (17.08) | |
| Comorbidity, before 1 year | |||
| Diabetes mellitus | |||
| Yes | 1358 (30.35) | 802 (17.93) | <.0001 |
| No | 3116 (69.65) | 3672 (82.07) | |
| Hypertension | |||
| Yes | 2576 (57.58) | 1608 (35.94) | <.0001 |
| No | 1898 (42.42) | 2866 (64.06) | |
| Autoimmune disease requiring steroid therapy | |||
| Yes | 76 (1.70) | 54 (1.21) | .0519 |
| No | 4398 (98.30) | 4420 (98.79) | |
The number of patients with at least 1 cataract surgery–related complication was not significantly different between the α 1 -Blocker group (385 [8.61%]) and the No α 1 -Blocker group (358 [8%]). The number of patients with at least 1 cataract reoperation–related complication was not significantly different between the α 1 -Blocker group (83 [1.86%]) and the No α 1 -Blocker group (72 [1.61%]). The 2 most common complications in both groups were glaucoma and a dropped nucleus, followed by vitreous hemorrhage, retinal detachment, and wound dehiscence in the α 1 -Blocker group; and followed by retinal detachment, vitreous hemorrhage, and endophthalmitis in the No α 1 -Blocker group. The differences in the number of individual complications were not significant between the 2 groups ( Table 2 ).
| α 1 -Blocker (n = 4474) | No α 1 -Blocker (n = 4474) | Odds Ratio (95% CI) | P Value | AOR a (95% CI) | P Value | |
|---|---|---|---|---|---|---|
| All complications [no. of patients (%)] | 385 (8.61) | 358 (8.00) | 0.97 (0.73–1.30) | .8535 | 0.85 (0.61–1.16) | .3022 |
| All complications for reoperations | 83 (1.86) | 72 (1.61) | 0.85 (0.44–1.67) | .6391 | 0.85 (0.41–1.75) | .6594 |
| No. of complications | 0.94 (0.75–1.18) | .6151 | 0.87 (0.68–1.12) | .2804 | ||
| No. of surgical complications | 0.87 (0.54–1.38) | .5466 | 0.87 (0.53–1.44) | .5912 | ||
| ICD-9-CM | ||||||
| Glaucoma | 286 (6.39) | 289 (6.46) | 0.90 (0.65–1.24) | .5105 | 0.78 (0.55–1.11) | .1693 |
| Ocular hypertension | 15 (0.34) | 26 (0.58) | 0.26 (0.05–1.25) | .0914 | 0.21 (0.04–1.12) | .0680 |
| Endophthalmitis | 10 (0.22) | 11 (0.25) | 0.59 (0.08–4.30) | .6053 | 0.68 (0.08–5.71) | .7213 |
| Vitreous hemorrhage | 56 (1.25) | 32 (0.72) | 1.66 (0.68–4.04) | .2665 | 1.39 (0.54–3.62) | .4971 |
| ICD-9-CM procedural code | ||||||
| Retinal detachment | 21 (0.47) | 20 (0.45) | 0.87 (0.19–3.94) | .8595 | 0.82 (0.16–4.36) | .8184 |
| Intraocular lens dislocation | 3 (0.07) | 9 (0.20) | 0.21 (0.02–1.85) | .1589 | 0.25 (0.03–2.32) | .2214 |
| Dropped nucleus | 68 (1.52) | 51 (1.14) | 0.86 (0.41–1.80) | .6875 | 0.79 (0.36–1.76) | .5706 |
| Pseudophakic corneal edema | 6 (0.13) | 11 (0.25) | 0.87 (0.05–15.38) | .9217 | 1.37 (0.08–22.30) | .8249 |
| Wound dehiscence | 11 (0.25) | 8 (0.18) | 1.73 (0.28–10.55) | .5512 | 2.14 (0.34–13.54) | .4208 |
a AOR adjusted by age, sex, urbanization, hospital level, and comorbidities.
We also identified 922 patients in the α 1 -Blocker group who had taken tamsulosin before the surgery. They were older (not significant), were primarily male (93%; P < .0001), and had significantly higher percentages of hypertension, diabetes (both P < .0001), and steroid-treated autoimmune disease ( P = .0057) ( Table 3 ).
| Tamsulosin (n = 922) | No α 1 -Blocker (n = 4474) | P Value | |
|---|---|---|---|
| Age (y) (mean ± SD) | 72.35 ± 7.76 | 71.83 ± 7.92 | .0673 |
| Age group (%) | |||
| ≤65 | 142 (15.40) | 766 (17.12) | .2037 |
| >65 | 780 (84.60) | 3708 (82.88) | |
| Sex (%) | |||
| Female | 64 (6.94) | 1020 (22.80) | <.0001 |
| Male | 858 (93.06) | 3454 (77.20) | |
| Medical level (%) | |||
| Medical center | 218 (23.64) | 968 (21.64) | .0002 |
| Regional hospital | 178 (19.31) | 656 (14.66) | |
| Local hospital | 526 (57.05) | 2850 (63.70) | |
| Urbanization | |||
| 1 | 236 (25.60) | 1155 (25.82) | .9484 |
| 2 | 438 (47.51) | 2097 (46.87) | |
| 3 | 97 (10.52) | 458 (10.24) | |
| 4 | 151 (16.38) | 764 (17.08) | |
| Comorbidity, before 1 year | |||
| Diabetes mellitus | |||
| Yes | 276 (29.93) | 802 (17.93) | <.0001 |
| No | 646 (70.07) | 3672 (82.07) | |
| Hypertension | |||
| Yes | 482 (52.28) | 1608 (35.94) | <.0001 |
| No | 440 (47.72) | 2866 (64.06) | |
| Autoimmune disease requiring steroid therapy | |||
| Yes | 22 (2.39) | 54 (1.21) | .0057 |
| No | 900 (97.61) | 4420 (98.79) | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
