Postoperative adverse outcomes in surgical patients with epilepsy: A population-based study

Authors

  • Chuen-Chau Chang,

    1. Department of Anesthesiology, Affiliated with Health Policy Research Center, Taipei Medical University Hospital, Taipei, Taiwan
    2. Department of Anesthesiology, Taipei Medical University, Taipei, Taiwan
    Search for more papers by this author
  • Chaur-Jong Hu,

    1. Department of Neurology, Taipei Medical University, Taipei, Taiwan
    Search for more papers by this author
  • Fai Lam,

    1. Department of Anesthesiology, Affiliated with Health Policy Research Center, Taipei Medical University Hospital, Taipei, Taiwan
    Search for more papers by this author
  • Hang Chang,

    1. Department of Emergency Medicine, Shin Kong Wu Ho-Su Memorial Hospital, Taipei, Taiwan
    Search for more papers by this author
  • Chien-Chang Liao,

    1. Department of Anesthesiology, Affiliated with Health Policy Research Center, Taipei Medical University Hospital, Taipei, Taiwan
    2. Department of Anesthesiology, Taipei Medical University, Taipei, Taiwan
    Search for more papers by this author
  • Ta-Liang Chen

    1. Department of Anesthesiology, Affiliated with Health Policy Research Center, Taipei Medical University Hospital, Taipei, Taiwan
    2. Department of Anesthesiology, Taipei Medical University, Taipei, Taiwan
    Search for more papers by this author

  • Chien-Chang Liao is also equal as first author.

Address correspondence to Ta-Liang Chen, Department of Anesthesiology, Taipei Medical University Hospital, 252 Wuxing St., Taipei 11031, Taiwan. E-mail: tlc@tmu.edu.tw

Summary

Purpose:  People with epilepsy are more likely than healthy people to experience comorbidities and complications in various medical situations. However, the prevalence of postoperative complications, mortality, and use of medical resources in surgical patients with epilepsy has not been studied. The purpose of this study is to examine whether epilepsy is an independent risk factor for postoperative adverse outcomes of patients receiving major surgery.

Methods:  Retrospective cohort study using the National Health Insurance Research Database to identify patients with epilepsy who underwent major surgery in Taiwan between the years 2004 and 2007. For each case, four age- and sex-matched participants without epilepsy were included. Preoperative comorbidities in the 24 months before surgery were identified. Eight major postoperative complications, overall 30-day mortality, and in-hospital utilization of medical resources (including length of hospital stay, percentage of postoperative intensive care unit admissions, and in-hospital medical expenditures) served as the major outcome measurements. Comorbidities, status of receiving renal dialysis, teaching hospital status, types of surgery, and patients living in urban or rural areas were adjusted by multivariate logistic regression.

Key Findings:  A total of 13,103 participants with epilepsy and 52,412 without were included. Patients with epilepsy have significantly more preoperative comorbidities and demonstrated more risks of any postoperative complications (odds ratio 2.02, 95% confidence interval 1.90–2.14). Consumption of in-hospital medical resources was also significantly higher in patients with epilepsy, but no significant differences in postoperative mortality rates between the two groups were noted.

Significance:  Stroke was identified as the most significant postoperative complication for surgical patients with epilepsy. Patients, especially those with previous hospitalization or emergency visits due to the disease, confronted significantly higher postoperative complication rates, and consumed more in-hospital medical resources without differences in overall mortality rates. Further revision of health care standards to provide early recognition of postoperative complications and better management for surgical patients with epilepsy is needed.

Epilepsy is the most common serious neurologic disorder; it affects an estimated 50 million people worldwide (Duncan et al., 2006). Epilepsy can occur at all ages with different clinical presentations and causes (Brodie & French, 2000). Previous research on patients with epilepsy receiving surgery has focused mainly on coexisting medical illnesses and perioperative recurrence of seizures (Jalava & Sillanpää, 1996; Niesen et al., 2010). Earlier studies on postoperative complications and mortality were limited only to specific procedures (Niesen et al., 2010; McClelland et al., 2011), and the knowledge of global features of perioperative adverse outcomes in surgical patients with epilepsy was still lacking. To offer effective and high-quality health care, we need to define a comprehensive understanding of surgery-related complications and mortalities in patients with epilepsy (Kelley et al., 2009). On the other hand, cost-of-illness (COI) studies of epilepsy have illustrated that the illness is a substantial burden both to individuals and to society (De Zélicourt et al., 2000; Pugliatti et al., 2007). Estimation of medical use of in-hospital medical resources (including length of hospital stay, percentage of postoperative intensive care unit (ICU) admission, and in-hospital medical expenditures) of surgical patients with epilepsy is crucial for health care resources allocation and could have implications for policy change.

This study seeks to delineate the comorbidities, postoperative complications and mortality, and use of in-hospital medical resources among surgical patients with epilepsy. Our hypothesis is that patients with epilepsy might present with increased risk of complications after surgery.

Material and Methods

Source of data

This study used the reimbursement claims data of the Taiwan National Health Insurance Program, which was reformed from 13 prior health insurance schemes and started in March 1995. With universal coverage, >99% of 22.6 million Taiwan residents have been enrolled in this system. The Taiwan National Health Research Institute collects information on all beneficiaries in the National Health Insurance Research Database (NHIRD), including patient demographics, primary and secondary diagnoses of disease, procedures, prescriptions, and medical expenditures. The institute also records all inpatient and outpatient use of medical services for all beneficiaries. To protect patients’ privacy, the electronic database was decoded with patient identifications scrambled for further public access. The study was evaluated and approved by the NHIRD research committee.

Study population

By using NHI procedure codes (http://www.nhi.gov.tw/), we examined medical claims and identified 13,103 surgical patients from among 2,010,412 persons who underwent inpatient major surgeries (defined as surgeries requiring general, epidural, or spinal anesthesia and hospitalized for more than 1 day) from 2004–2007 and had a diagnosis of epilepsy, which was defined by the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) code 345. To increase the likelihood of capturing true positive cases with epilepsy, at least two claims coded as epilepsy in their primary diagnosis during the 24-month period before the index surgical admission were required. For each surgical patient with epilepsy, we randomly selected four surgical patients matched for sex and age who had no preoperative history of epilepsy. Patients receiving epilepsy surgeries were excluded from this study.

Measures

Preoperative major coexisting medical illnesses were recorded from medical claims for a 24-month period preoperatively; these were acute myocardial infarction, acute renal failure, chronic obstructive pulmonary disease, congestive heart failure, diabetes mellitus, hypertension, peripheral vascular disease, mental disorders, traumatic brain injury, and stroke. Similarly to epilepsy, comorbidities were defined as those preoperative conditions coded as primary diagnosis for at least two claims of ambulatory and/or inpatient medical care during the 24-month preoperative period. A history of receiving renal dialysis was considered as preoperative functional status. Eight postoperative conditions were examined: acute myocardial infarction, acute renal failure, deep wound infection, pneumonia, postoperative bleeding, pulmonary embolism, septicemia, and stroke (Khuri et al., 2005; Ghaferi et al., 2009). Postoperative complications were defined as those postoperative conditions occurring within the 30-day periods after the index surgeries, with special limitation to exclude the presence of these conditions in the 24-month preoperative periods. These complications and subsequent overall in-hospital mortality within 30 days after index surgery were the study’s primary outcomes. We defined comorbidities and postoperative complications as acute myocardial infarction (ICD-9-CM 410), acute renal failure (ICD-9-CM 584), chronic obstructive pulmonary disease (ICD-9-CM 490–496), congestive heart failure (ICD-9-CM 428), deep wound infection (ICD-9-CM 958), diabetes mellitus (ICD-9-CM 250), hypertension (ICD-9-CM 401–405), peripheral vascular disease (ICD-9-CM 443), pneumonia (ICD-9-CM 480–486), postoperative bleeding (ICD-9-CM 998.0, 998.1, and 998.2), pulmonary embolism (ICD-9-CM 415), septicemia (ICD-9-CM 038 and 998.5), mental disorders (ICD-9-CM 290–319), traumatic brain injury (ICD-9-CM 800–804 and 850–854), and stroke (ICD-9-CM 430–438).

Also considered as parameters of medical resources use were whether the surgery was performed in a teaching hospital, types of surgery, length of stay, and use of ICU and in-hospital medical expenditures. Types of surgery included skin, breast, musculoskeletal, respiratory, cardiovascular, eye, digestive, kidney, ureter, bladder, delivery, cesarean section, abortion, neurosurgery, and others. In-hospital medical expenditure is the total medical costs estimated by the sum of all cost components including the costs of diagnosis, ward admission, examination, radiation therapy, medical treatment, surgery, rehabilitation, blood transfusion, anesthesia, medical material, medication, psychiatric specific treatment, injection, and personal copayments. Population densities (persons/km2) calculated by dividing the population (persons) by the area (km2) for each administrative unit of Taiwan were categorized into four quartiles as low, moderately, highly, and very highly urbanized areas (Shih et al., 2010). Low income status was defined as patients qualified for waiving medical copayments, a qualification certified by the Bureau of Taiwan National Health Insurance.

To explore more clinical risk factors predicting postoperative adverse outcomes for preoperative epilepsy, previous histories including hospitalization and emergency visits for epilepsy care within the 24-month period before the index surgery were also considered for further analysis.

Statistical analysis

Using chi-square tests, descriptive analyses concerning distribution of demographic status, coexisting medical conditions, types of surgery, use of inpatient medical resources (including length of hospital stay, percentage of postoperative ICU admission, and in-hospital medical expenditures), and postoperative complication and mortality rates between surgical patients with and without epilepsy were compared.

Odds ratios (ORs) with 95% confidence intervals (CIs) for 30-day postoperative mortality, and complications between patients with and without epilepsy were analyzed using multivariate logistic regression to adjust for operation in teaching hospital or not, low-income status, receiving dialysis or not, urbanization, preoperative coexisting medical conditions, types of surgery, and ICU stay. Each coexisting disease was considered as a single variable entered into the logistic regression model separately. Whether postoperative complications were associated with the epilepsy was further analyzed by preoperative diagnosis and history of emergency visits or hospitalization due to epilepsy and controlled other covariates with multivariate logistic regression. SAS software version 9.1 (SAS Institute Inc., Cary, NC, U.S.A.) was used for data analyses with two-sided probability, and a p-value <0.05 was considered statistically significant.

Results

This study included 13,103 patients with a preoperative diagnosis of epilepsy and 52,412 patients with no history of epilepsy (Table 1). Patients with epilepsy had more surgeries performed in teaching hospitals, and tended to be of lower income status and to live in relatively less urbanized areas compared with patients without epilepsy. There were more coexisting medical conditions in epilepsy cases than in controls. Although patients with epilepsy tended to have fewer digestive, urologic, and gynecological procedures, they had more neurologic procedures than patients without epilepsy, especially intracranial, skull, scalp, and extracranial operations.

Table 1.   Preoperative characteristics of surgical patients with and without epilepsy
 Without epilepsy
N = 52,412
no. (%)
With epilepsy
N = 13,103
no. (%)
p-value
  1. COPD, chronic obstructive pulmonary disease; CS, cesarian section.

Sex  1.00
 Female20,780 (39.7)5,195 (39.7) 
 Male31,632 (60.3)7,908 (60.3) 
Age (years)  1.00
 20–296,352 (12.1)1,588 (12.1) 
 30–397,180 (13.7)1,795 (13.7) 
 40–499,436 (18.0)2,359 (18.0) 
 50–598,548 (16.3)2,137 (16.3) 
 60–697,468 (14.3)1,867 (14.3) 
 ≥7013,428 (25.6)3,357 (25.6) 
 Mean ± SD54.1 ± 18.454.3 ± 18.60.2707
Operation in teaching hospital  <0.0001
 No6,690 (12.8)1,315 (10.0) 
 Yes45,722 (87.2)11,788 (90.0) 
Low-income  <0.0001
 No51,135 (97.6)11,804 (90.1) 
 Yes1,277 (2.4)1,299 (9.9) 
Urbanization  <0.0001
 Low12,444 (23.7)3,627 (27.7) 
 Moderate12,135 (23.2)3,114 (23.8) 
 High14,464 (27.6)3,348 (25.6) 
 Very high13,369 (25.5)3,014 (23.0) 
Coexisting medical conditions   
 Stroke5,062 (9.7)5,089 (38.8)<0.0001
 Acute renal failure180 (0.3)110 (0.8)<0.0001
 Peripheral vascular disease618 (1.2)220 (1.7)<0.0001
 Congestive heart failure1,810 (3.5)687 (5.2)<0.0001
 Myocardial infarction5,114 (9.8)1,654 (12.6)<0.0001
 COPD10,695 (20.4)3,676 (28.1)<0.0001
 Diabetes10,274 (19.6)3,343 (25.5)<0.0001
 Hypertension15,617 (29.8)4,509 (34.4)<0.0001
 Dialysis600 (1.1)316 (2.4)<0.0001
 Traumatic brain injury3,190 (6.1)2,859 (21.8)<0.0001
 Mental disorders10,261 (19.6)5,806 (44.3)<0.0001
Types of surgery  <0.0001
 Skin2,097 (4.0)618 (4.7) 
 Breast737 (1.4)129 (1.0) 
 Musculoskeletal15,829 (30.2)4,090 (31.2) 
 Respiratory2,583 (4.9)721 (5.5) 
 Cardiovascular1,665 (3.2)472 (3.6) 
 Eye492 (0.9)130 (1.0) 
 Digestive11,725 (22.4)2,240 (17.1) 
 Kidney, ureter, bladder4,254 (8.1)761 (5.8) 
 Delivery, CS, abortion3,618 (6.9)542 (4.1) 
 Neurosurgery3,760 (7.2)2,404 (18.4) 
 Other5,652 (10.8)996 (7.6) 
Neurosurgical procedures  <0.0001
 Intracranial1,202 (32.0)1,686 (70.1) 
 Spine2,234 (59.4)459 (19.1) 
 Peripheral nerves190 (5.1)51 (2.1) 
 Skull, scalp, extracranial vessels134 (3.6)208 (8.7) 

Compared with patients without epilepsy, patients with epilepsy had a higher percentage of 30-day postoperative stroke, pneumonia, septicemia, acute renal failure, postoperative bleeding, deep wound infection, and overall complications (20.6% vs. 9.8%, p < 0.0001) (Table 2). The hospital length of stay for patients with epilepsy was longer than that for patients without epilepsy (17.8 ± 25.5 vs. 10.4 ± 14.8, p < 0.0001). ICU usage was higher by patients with epilepsy than patients without (30.1% vs. 13.5%, p < 0.0001). Higher in-hospital medical expenditures were found in surgical patients with epilepsy than in patients without epilepsy. Overall 30-day postoperative mortality was 1.1% for patients with epilepsy, which was much higher than 0.6% for patients without epilepsy (p < 0.0001).

Table 2.   Postoperative characteristics of surgical patients with and without epilepsy
 Without epilepsy N = 52,412
no. (%)
With epilepsy
N = 13,103
no. (%)
p-value
  1. ICU, intensive care unit; ME, medical expenditure; USD, US dollars.

  2. aMean (SD).

Postoperative complications   
 Stroke848 (1.6)531 (4.1)<0.0001
 Pneumonia1,490 (2.8)1,262 (9.6)<0.0001
 Septicemia1,238 (2.4)833 (6.4)<0.0001
 Acute renal failure427 (0.8)196 (1.5)<0.0001
 Postoperative bleeding1,686 (3.2)469 (3.6)0.0374
 Acute myocardial infarction159 (0.3)37 (0.3)0.6940
 Deep wound infection341 (0.7)126 (1.0)0.0002
 Pulmonary embolism52 (0.1)11 (0.1)0.6141
 Any of the above5,140 (9.8)2,705 (20.6)<0.0001
Length of stay, days  <0.0001
 1–525,521 (48.7)4,198 (32.0) 
 6–1013,032 (24.9)2,972 (22.7)
 11–155,041 (9.6)1,575 (12.0)
 >158,818 (16.8)4,358 (33.3)
 Median (interquartile range)6 (8)9 (18)<0.0001
ICU stay, %7,085 (13.5)3,940 (30.1)<0.0001
In-hospital ME, USDa   
 Low795 (142.1)908 (197.9)<0.0001
 Medium1,245 (188.2)1,763 (305.5)<0.0001
 High2,250 (514.6)3,962 (1045.1)<0.0001
 Very high9,081 (7701.0)14,141 (10097.3)<0.0001
In-hospital mortality312 (0.6)148 (1.1)<0.0001

After adjusting for teaching hospital status, low income status, urbanization, coexisting medical conditions, types of surgery, and ICU stay, the significant ORs between the groups for pneumonia, stroke, septicemia, acute renal failure, deep wound infection, postoperative bleeding, and overall complications were 2.54, 3.15, 2.03, 1.61, 1.31, 1.14, and 2.02, respectively (Table 3). However, the OR of 30-day postoperative mortality for patients with epilepsy compared to those without was 1.11 (95% CI 0.89–1.40) without significant difference after adjustment. (Data not shown in tables.)

Table 3.   Preoperative epilepsy associated with postoperative complications in multiple logistic regression modelsa
 Preoperative epilepsy
None
N = 52,412
Epilepsy
N = 13,103
OR (95% CI)OR (95% CI)
  1. OR, odds ratio; CI, confidence interval.

  2. aAdjusted for age, sex, surgery in teaching hospital, low-income status, receiving dialysis or not, urbanization, comorbidities, types of surgery, and ICU stay.

Postoperative complications  
 Pneumonia1.00 (reference)2.54 (2.32–2.79)
 Stroke1.00 (reference)3.15 (2.76–3.59)
 Septicemia1.00 (reference)2.03 (1.83–2.26)
 Acute renal failure1.00 (reference)1.61 (1.32–1.97)
 Deep wound infection1.00 (reference)1.31 (1.04–1.66)
 Postoperative bleeding1.00 (reference)1.14 (1.02–1.29)
 Any of the above1.00 (reference)2.02 (1.90–2.14)

Table 4 shows the associations between postoperative complications and both preoperative hospitalization or emergency visits for epilepsy care. Compared with those without epilepsy, epilepsy cases with history of hospitalization for epilepsy during the 24-month period before surgery had higher ORs of developing postoperative pneumonia (3.54), stroke (4.92), septicemia (2.71), acute renal failure (2.27), deep wound infection (1.44), and any postoperative complications (2.69). However, epilepsy cases with history of emergency visits for epilepsy presented increased ORs for all these complications except acute renal failure.

Table 4.   The association between preoperative history of hospitalization or emergency visits for epilepsy and postoperative complicationsa
 History of hospitalization for epilepsyHistory of emergency visits for epilepsy
Control N = 52,412Case without history
N = 7,811
Case with history N = 5,292Control N = 52,412Case without history N = 10,080Case with history N = 3,023
OR (95% CI)OR (95% CI)OR (95% CI)OR (95% CI)OR (95% CI)OR (95% CI)
  1. OR, odds ratio; CI, confidence interval.

  2. aAdjusted for age, sex, surgery in teaching hospital, low-income status, receiving dialysis or not, urbanization, comorbidities, types of surgeries, and intensive care unit stay.

Postoperative complications      
 Pneumonia1.00 (reference)1.92 (1.72–2.16)3.54 (3.16–3.97)1.00 (reference)2.50 (2.26–2.75)2.73 (2.34–3.18)
 Stroke1.00 (reference)2.23 (1.89–2.63)4.92 (4.17–5.81)1.00 (reference)3.00 (2.60–3.45)3.73 (3.01–4.63)
 Septicemia1.00 (reference)1.61 (1.41–1.83)2.71 (2.38–3.09)1.00 (reference)2.03 (1.82–2.27)2.05 (1.71–2.46)
 Acute renal failure1.00 (reference)1.21 (0.94–1.55)2.27 (1.78–2.90)1.00 (reference)1.59 (1.29–1.97)1.71 (1.19–2.45)
 Deep wound infection1.00 (reference)1.22 (0.92–1.63)1.44 (1.05–1.95)1.00 (reference)1.15 (0.88–1.50)1.82 (1.29–2.56)
 Postoperative bleeding1.00 (reference)1.17 (1.02–1.34)1.11 (0.94–1.31)1.00 (reference)1.15 (1.01–1.30)1.13 (0.91–1.39)
 Any of above1.00 (reference)1.63 (1.52–1.76)2.69 (2.49–2.91)1.00 (reference)1.99 (1.86–2.12)2.13 (1.92–2.36)

Discussion

Major findings

In this population-based study, we demonstrated that surgical patients with a preoperative diagnosis of epilepsy have significantly higher prevalence of somatic comorbidities of stroke, acute renal failure, peripheral vascular disease, congestive heart failure, myocardial infarction, chronic obstructive pulmonary disease, diabetes, and hypertension. Compared with surgical patients without epilepsy, those with epilepsy exhibited higher 30-day postoperative complication rates for stroke, pneumonia, septicemia, acute renal failure, bleeding, and deep wound infection, with significantly higher in-hospital utilization of medical resources including length of stay, postoperative ICU stay, and medical expenditures. With the highest OR of 3.15, stroke was identified as the most significant postoperative complication. However, there is no significant difference in overall mortality rate between the two groups after adjustment for all covariates.

Comparison with previous studies

Most previous community-based studies on patients with epilepsy focused only on psychiatric and psychosomatic comorbidities (Oyegbile et al., 2004; Pellock, 2004; Danielsson et al., 2005; Christensen et al., 2007; Gonzalez-Heydrich et al., 2007; Adams et al., 2008; Lothe et al., 2008; Stefanello et al., 2010). Research concerning somatic comorbidities in epilepsy revealed its correlations with hypertension, anemia (Lee et al., 2005), peptic ulcers, urinary incontinence, bowel disorders, heart diseases, chronic bronchitis/emphysema (Téllez-Zenteno et al., 2005), pneumonia, fractures, cardiovascular and cerebrovascular disorders (Gaitatzis et al., 2004), and diabetes (Hinnell et al., 2010). Most previous studies on such postoperative complications focused on specific procedures with limited outcome measurements without the global features for surgical patients with epilepsy (Arozullah et al., 2001; Moretti et al., 2005; Gutsche et al., 2007; Bateman et al., 2009; Kheterpal et al., 2009; McClelland et al., 2011). In contrast with previous studies (Gerson et al., 1990; Reilly et al., 1999; Bamgbade et al., 2007), we verified that surgical patients with epilepsy exhibited significantly increased postoperative complications across a full spectrum compared with those without epilepsy.

On the other hand, few studies had delineated the profile of surgical procedures in patients with epilepsy. In the current study, we demonstrated that patients with epilepsy had more neurologic procedures than those without epilepsy, especially intracranial, skull, scalp, and extracranial operations. This is compatible with a priori understanding of the risk factors for epilepsy, including head injuries, stroke and other vascular diseases, and infections (Velioğlu et al., 2001; Lowenstein, 2009; van Baalen et al., 2010). However, further exploration is needed to establish and rationalize the observation of having fewer procedures in digestive, urologic, and gynecologic procedures for patients with epilepsy.

Previous studies had demonstrated that the onset of seizures in late life is associated with a striking increase in the risk of stroke (Cleary et al., 2004). In a most recent study, this increased risk of stroke was suggested to be associated with the use of antiepileptic drugs (AEDs) (Olesen et al., 2011). With epilepsy patients demonstrating this baseline vulnerability for stroke, our study is compatible with those previous works, and demonstrates increased risk of stroke in surgical settings. We postulated that AED turbulence might play some roles in this scenario. However, this hypothesis deserves further exploration.

Plausible mechanisms

Previous studies have demonstrated increased long-term mortality in nonselected patients with epilepsy (Cockerell et al., 1994; Lhatoo et al., 2001), suggesting increased risk of premature death either from the underlying illnesses or from epilepsy (Cockerell et al., 1994). Our study found no significant difference between surgical patients with and without epilepsy in overall postoperative mortality after adjusting for multiple covariates. It is interesting to find increased overall perioperative complication rates, whereas mortality rates were unaffected in surgical patients with epilepsy. Because long-term cohort studies have demonstrated diverse somatic comorbidities in patients with untreated epilepsy (Cockerell et al., 1994; Lhatoo et al., 2001; Nicoletti et al., 2009), we postulate that subclinical comorbidities might emerge as major complications after surgical challenges. Another factor is pharmacologic. A recent study demonstrated that patients with epilepsy might be exposed to complex pharmacotherapy of two to eight concomitant drugs (Karouni et al., 2010). Polypharmacy for epilepsy or for its comorbidities might contribute to complications caused side effects or drug-to-drug interactions thus generated (Patsalos et al., 2002). A third factor is that many commonly used drugs (such as analgesics, anesthetics, and drugs for stroke patients) generally disturb spike-wave discharges (van Luijtelaar et al., 2002). Therefore, anesthetics and analgesics might complicate postoperative conditions in patients with epilepsy by blunting responses to mild derangement. This communication disturbance consistently leads to delays in seeking medical assistance. Together with nonspecific and misleading presenting symptoms and signs of complications, these factors might delay diagnosis and treatment.

On the other hand, once a diagnosis of postoperative complications is established, a medical history of epilepsy including previous hospitalization or emergency visits due to epilepsy might trigger a more extensive and comprehensive management plan, and evidenced by increased utilization of in-hospital medical resources. A perceived need for integrated care might thus help reduce expected mortality. Similar features were observed in cohorts of poor self-reported exercise tolerance or patients with Parkinsonism with increased postoperative complication rates with comparable mortality (Simon et al., 1992; Reilly et al., 1999).

Most previous cost-of-illness (COI) studies on epilepsy described disease burden (Frost et al., 2000; Pugliatti et al., 2007; Hong et al., 2009), factors affecting the COI (Lee et al., 2005; Davis et al., 2008), and related medical expenditures in nonselected patients with epilepsy (van den Broek et al., 2004; Morgan & Kerr, 2004). Our data demonstrated longer hospital stays, more ICU stays, and higher in-hospital medical expenditures for patients with epilepsy in surgical settings. Epilepsy might increase in-hospital expenditures in several ways. First, the inpatient care for the disease itself costs more than care for the general population (Morgan & Kerr, 2004). The presence of comorbidities mandates more complex and integrated care (Lee et al., 2005), and the use of AEDs might also contribute notably (Hamer et al., 2006). Secondly, significantly higher ICU admission rates among surgical patients with epilepsy serve to increase in-hospital expenditures for this group. However, the correlation between in-hospital expenditures and overall postoperative survival needs further clarification.

Strength and limitations

The statistical power of this population-based study was sufficient to delineate preoperative comorbidity and postoperative complications. There are still some limitations that need to be addressed.

NHIRD data structure

Among the limitations, the reimbursement claims database is lacking in details regarding patient characteristics, specific etiology of epilepsy, and body mass index that were related with patients’ comorbidities, complications, mortality, and medical expenditures (Cockerell et al., 1994; De Zélicourt et al., 2000; Lhatoo et al., 2001; Kheterpal et al., 2007; Daniels et al., 2009). In the NHIRD, one major diagnosis with two secondary diagnoses were recorded for outpatient claims and one major with four secondary diagnoses for inpatient claims was collected, respectively. With preoperative observing periods expanding to 24 months, we believed that the possibility of missing any significant comorbidity is low. However, this data structure might limit the number of captured postoperative complications to some extent.

Intraoperative parameters

Our database did not include such detailed parameters as physical status classification according to the American Society of Anesthesiologists (ASA) as well as body temperature, blood pressure, and duration for surgical procedures. There were also differences between the profile of surgical procedures, with epilepsy patients having fewer digestive, urologic, and gynecologic procedures, but more neurologic operations. These might affect complications and/or mortality to some extent (Ziser et al., 1999; Abelha et al., 2009). Although we have adjusted coexisting medical illnesses as a surrogate for ASA classification, urbanization and teaching hospital status for technical variations, and also adjusted types of surgery, this nonrandomization could exert residual influence over outcomes and deserves further inspection.

Anesthetic techniques

A third limitation was that we did not incorporate variables concerning anesthetic techniques into the statistical estimation. Although anesthetic techniques were documented as not affecting risk of perioperative seizures in patients with a history of epilepsy (Niesen et al., 2010), these techniques have been clearly associated with perioperative complication and mortality (Röhm et al., 2006; Wohlrab et al., 2009). We believed that, with such a large-scaled sample in our current study, the randomly distributed anesthetic techniques between groups might not have biased the estimated ORs.

Capturing strategy

We used medical records and ICD-9-CM codes to define epilepsy cases, controls, comorbidities, and complications in the present study. With our present capturing strategy comparable to the guidelines developed by the International League Against Epilepsy (ILAE) (Commission on Epidemiology and Prognosis, ILAE, 1993), chances for misclassifying stroke-related acute seizures as epilepsy cases are slim. However, it is possible that the controls had epilepsy but that their codes predated the 24-month period, rendering the current data a conservative estimate of the surgical complications. The National Health Insurance Bureau administers regular sample cross-checks of each hospital’s claims with medical charts, and punitive measures for coding infractions, followed by tying a hospital’s reimbursement level to its patient severity profile. Although hospitals’ interests are best served by accurate coding of diagnoses and care items, and the precision of procedure codes in NHIRD is generally recognized, miscoding by clinical physicians may still occur as a result of human error (Lin et al., 2008). Most recent study demonstrated the accuracy of major diagnosis codes in NHIRD; however, validity of comorbidity and complication codes might still be one of the study limitations (Cheng et al., 2011).

Causality

Finally, although population-based health care administrative data have been used for seizure-related studies, and our research did demonstrate the correlation between preoperative epilepsy and postoperative complications, causality cannot be inferred until further prospective studies (Kozyrskyj & Prasad, 2004).

Conclusion

We identified that surgical patients with epilepsy had comorbidities including various medical illnesses, and faced significantly higher postoperative complication rates in stroke, pneumonia, septicemia, and acute renal failure. These comorbidities and complications were closely related to epilepsy, so surgical patients with epilepsy consumed more in-hospital medical resources even though their mortality rates were similar to surgical patients without epilepsy.

The policy implications of this study suggest the need for several changes, including a disease-oriented health care standard for surgical patients with epilepsy. These results also suggest that the knowledge and alertness to the postoperative complications can facilitate appropriate treatment. Such measures can contribute to better assessment and management of surgical patients with epilepsy.

Acknowledgments

Research support was provided by Taipei Medical University institutional and/or departmental sources.

Disclosure

None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Ancillary