Abstract
OBJECTIVE:
For adult patients undergoing surgical decompression for Chiari malformation type I (CM-I), the patient-level factors that influence extended length of stay (LOS) are relatively unknown. The aim of this study was to investigate the impact of patient-baseline comorbidities, demographics, and postoperative complications on extended LOS after intervention after adult CM-I decompression surgery.
METHODS:
A retrospective cohort study using the National Inpatient Sample years 2010–2014 was performed. Adults (≥18 years) with a primary diagnosis of CM-I undergoing surgical decompression were identified. Weighted patient demographics, comorbidities, complications, LOS, disposition, and total cost were recorded. A multivariate logistic regression was used to determine the odds ratio for risk-adjusted LOS.
RESULTS:
A total of 29,961 patients were identified, 6802 of whom (22.7%) had extended LOS. The extended LOS cohort had a significantly greater overall complication rate (normal LOS, 10.6% vs. extended LOS, 29.1%; P < 0.001) and total cost (normal LOS, $14,959 ± $6037 vs. extended LOS, $25,324 ± $21,629; P < 0.001) compared with the normal LOS cohort. On multivariate logistic regression, black race, income quartiles, private insurance, obstructive hydrocephalus, lack of coordination, fluid and electrolyte disorders, and paralysis were all independently associated with extended LOS. Additional duraplasty (P = 0.132) was not significantly associated with extended LOS after adjusting for other variables. The odds ratio for extended LOS was 2.07 (95% confidence interval, 1.59–2.71) for patients with 1 complication and 9.47 (95% confidence interval, 5.86–15.30) for patients with >1 complication.
CONCLUSIONS:
Our study shows that extended LOS after adult CM-I decompression surgery may be influenced by multiple patient-level factors.
Keywords: Chiari I malformation, Extended length of stay, Suboccipital decompression
INTRODUCTION
With the increase in health care costs in recent years, policy makers and hospitals are looking for ways to improve quality of health care and to reduce costs.1,2 One area that has gained national attention has been reducing length of stay (LOS).3,4 Because of the complexity of neurosurgical procedures, there is an increased predisposition for these patients to have extended LOS compared with other surgical and medical patients.5,6 Therefore, identifying risk factors that are associated with prolonged LOS after neurosurgical procedures is essential to creating avenues to improve patient quality of care and reduce health care costs.
Chiari malformation type I (CM-I) is a common neurosurgical diagnosis that is defined by a cerebellar tonsil ≥5 mm below the foramen magnum.2,7,8 The sequelae of this tonsillar herniation include headaches, neck pain, and numbness/tingling.9,10 Although new medical therapies are being explored to help patients with these symptoms, the overall treatment is surgical decompression of the foramen magnum.9 Although the prevalence of CM-I is increasing, so is the rate of surgical decompression. In a 14-year national trend analysis of the prevalence and operative rates of patients with CM-I in the United States, Wilkinson et al.11 found that the rate of surgical decompressions in adults increased 51% from the first half to the second half of the study period among younger patients and increased 28% among adult patients between 20 and 65 years of age. Although previous studies have looked at factors influencing surgical outcomes in adults, few studies have identified factors associated with extended LOS among these patients.
The aim of this study was to investigate the impact of patient-baseline comorbidities, demographics, and postoperative complications on extended LOS after intervention after adult CM-I decompression surgery.
METHODS
Data Source and Patient Population
The Healthcare Cost and Utilization Project’s National Inpatient Sample (NIS) database is a stratified discharge database representing 20% of all inpatient admissions from nonfederal hospitals in the United States. It is the largest all-payer health care database in the United States, containing approximately 7–8 million hospital admissions per year. A retrospective study was performed using years 2010–2014 of the NIS for all adult inpatient admissions undergoing surgical decompression for CM-I.
The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis and procedural coding system was used to identify patients and their respective comorbidities and surgical interventions. Adult patients (≥18 years old) with a primary diagnosis of CM-I (ICD-9-CM 348.4) were identified. ICD-9-CM procedural codes were then crossmatched to identify patients in the cohort undergoing either cranial decompression (ICD-9-CM 01.24) or spinal decompression of laminectomy (ICD-9-CM 03.09). This ICD-9-CM coding algorithm has previously been validated for patients undergoing CM-I decompression with good positive predictive value and sensitivity in administrative databases.12 Patients were then dichotomized according to whether they encountered an extended postoperative hospital LOS, which was defined as LOS greater than the 75th percentile for the entire cohort (>4 days).
Data Collection
Patient demographic information, comorbidities, and treating hospital characteristics were collected. Demographic information included age, gender, race, patient median household income quartile, and primary expected payer. Hospital characteristics included the region of the hospital, size by bed volume, and teaching status. Elixhauser comorbidities were used to evaluate incidence of deficiency anemias, rheumatoid arthritis/collagen vascular diseases, chronic pulmonary disease, coagulopathy, depression, diabetes, hypertension, hypothyroidism, fluid and electrolyte disorders, obesity, and paralysis. Presence of syringomyelia, hydromyelia, cervicalgia, scoliosis, headache, nausea and vomiting, obstructive hydrocephalus, muscle spasm, lack of coordination, and asthma were also assessed. Data on duraplasty, nerve/cord injury, and blood transfusion were included.
Complications for each admission were collected by indexing additional diagnoses. Complications assessed for included cerebral edema, dysphagia, acute postoperative pain, wound complication, sepsis, renal complication, urinary tract infection, acute respiratory complication, cardiac complications, genitourinary complications, deep vein thrombosis, acute posthemorrhagic anemia, central nervous system complication, and gastrointestinal complication. We then assessed patient outcomes of discharge disposition stratified by routine (home), nonroutine (short-term hospital + skilled nursing facility/acute rehabilitation + home with health care services), and other (leaving against medical advice–refusal of continued care, discharged to court/law enforcement, died in hospital, or location unknown) and total cost of hospital admission. All-payer inpatient cost/charge ratios were used to convert total hospital charge to total cost of hospital care.
Statistical Analysis
Discharge weights accounting for the redesign of the NIS in 2012 were used to calculate national estimates. Parametric data were expressed as mean ± standard deviation and compared via a 2-sided independent t test. Nonparametric data were expressed as median (interquartile range) and compared via the Mann-Whitney U test. Nominal data were compared with the χ2 test. For our primary hypothesis, we fit univariate and multivariate logistic regression with extended LOS as the dependent variable. Backward stepwise multivariate logistic regression analysis was used to select variables in a final model, using 0.1 as entry and stay criteria. We forced the treatment variable of interest (i.e., duraplasty) and complication during admission into the model in view of the plausibility for confounding. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. A P value <0.05 was determined to be statistically significant. Statistical analysis was performed using R Studio version 1.1.383 (RStudio Inc., Boston, Massachusetts).
RESULTS
Patient Demographics and Comorbidities
From the NIS database, 29,961 patients (≥18 years old) were identified undergoing surgical decompression. There were 23,159 patients (77.3%) who had an LOS ≤ 4 days and 6802 patients (22.7%) who had an extended LOS (LOS >4 days) (Table 1). Patients with extended LOS were older (normal LOS, 36.7 ± 12.0 years vs. extended LOS, 38.8 ± 13.3 years; P = 0.001) with similar proportions of female patients (P = 0.435) (Table 1). There was a greater proportion of white patients in the normal LOS cohort (normal LOS, 73.5% vs. extended LOS,62.3%; P < 0.001), and more black (normal LOS, 12.6% vs. extended LOS, 20.0%; P < 0.001) and Hispanic (normal LOS,8.5% vs. extended LOS, 12.1%; P < 0.001) patients in the extended LOS cohort (Table 1). The extended LOS cohort had more patients in the bottom 0–25th median household income quartile (normal LOS, 24.9% vs. extended LOS, 32.6%; P < 0.001) (Table 1). More patients with Medicaid encountered extended LOS (normal LOS, 16% vs. extended LOS, 19.1%; P < 0.001) (Table 1). There were no differences between the cohorts in terms of hospital region (P = 0.169), hospital bedsize (P = 0.357), or teaching status (P = 0.377) (Table 1).
Table 1.
Patient Demographics and Hospital Characteristics
| Variables | Normal LOS (≤4 Days) (n = 23,159) | Extended LOS (>4 Days) (n = 6802) | P |
|---|---|---|---|
| Age (years) | |||
| Mean ± standard deviation | 36.7 ±12.0 | 38.8±13.3 | 0.001 |
| Median (interquartile range) | 35 (27—45) | 37 (29—47) | 0.003 |
| Female (%) | 81.9 | 80.7 | 0.435 |
| Race (%) | <0.001 | ||
| White | 73.5 | 62.3 | |
| Black | 12.6 | 20.0 | |
| Hispanic | 8.5 | 12.1 | |
| Other | 5.4 | 5.6 | |
| Median household income quartile (%) | <0.001 | ||
| 0—25th | 24.9 | 32.6 | |
| 26—50th | 25.4 | 22.5 | |
| 51—75th | 28.0 | 27.1 | |
| 76—100th | 21.6 | 17.8 | |
| Primary expected payer (%) | <0.001 | ||
| Medicare | 6.9 | 11.1 | |
| Medicaid | 16.0 | 19.1 | |
| Private insurance | 69.3 | 60.0 | |
| Other | 7.8 | 9.8 | |
| Hospital demographics | |||
| Hospital region (%) | 0.169 | ||
| Northeast | 17.1 | 16.5 | |
| Midwest | 24.1 | 20.1 | |
| South | 37.6 | 42.4 | |
| West | 21.2 | 21.0 | |
| Hospital bedsize (%) | 0.357 | ||
| Small | 8.3 | 6.4 | |
| Medium | 17.1 | 17.3 | |
| Large | 74.6 | 76.2 | |
| Hospital teaching status (%) | 0.377 | ||
| Rural | 0.8 | 1.4 | |
| Urban nonteaching | 21.9 | 22.6 | |
| Urban teaching | 77.3 | 76.0 | |
LOS, length of hospital stay.
Presenting symptoms and comorbidities were overall greater in the extended LOS cohort. Compared with the normal LOS cohort, the extended LOS cohort had significantly more patients presenting with syringomyelia (normal LOS, 15.1% vs. extended LOS, 20.5%; P < 0.001), obstructive hydrocephalus (normal LOS,0.8% vs. extended LOS, 4.6%; P < 0.001), lack of coordination (normal LOS, 0.7% vs. extended LOS, 1.8%; P = 0.005), asthma (normal LOS, 12.4% vs. extended LOS, 15.1%; P = 0.047), deficiency anemias (normal LOS, 4.6% vs. extended LOS, 9.3%; P < 0.001), chronic pulmonary disease (normal LOS, 14.9% vs. extended LOS, 18.5%; P = 0.017), coagulopathy (normal LOS,0.7% vs. extended LOS, 2.0%; P = 0.001), depression (normal LOS, 13.3% vs. extended LOS, 16.4%; P = 0.047) diabetes (normal LOS, 4.6% vs. extended LOS, 8.6%; P < 0.001), hypertension (normal LOS, 20.6% vs. extended LOS, 30.7%; P < 0.001), fluid and electrolyte disorders (normal LOS, 4.2% vs. extended LOS,15.1%; P < 0.001), obesity (normal LOS, 13.7% vs. extended LOS,19.8%; P < 0.001), and paralysis (normal LOS, 0.6% vs. extended LOS, 2.5%; P < 0.001) (Table 2). There were similar rates of hydromyelia (P = 0.685), cervicalgia (P = 0.312), scoliosis P = 0.300), headache (P = 0.072), nausea and vomiting (P = 0.135), muscle spasm (P = 0.359), rheumatoid arthritis/collagen vascular diseases (P = 0.136), and hypothyroidism (P = 0.800) (Table 2).
Table 2.
Admission Diagnoses and Patient Comorbidities
| Variables (%) | Normal LOS (≤4 Days) (n = 23,159) | Extended LOS (>4 Days) (n = 6802) | P |
|---|---|---|---|
| Syringomyelia | 15.1 | 20.5 | <0.001 |
| Hydromyelia | 0.7 | 0.6 | 0.685 |
| Cervicalgia | 2.2 | 2.9 | 0.312 |
| Scoliosis | 1.1 | 1.5 | 0.300 |
| Headache | 19.9 | 23.2 | 0.072 |
| Nausea and vomiting | 9.3 | 11.2 | 0.135 |
| Obstructive hydrocephalus | 0.8 | 4.6 | <0.001 |
| Muscle spasm | 2.1 | 2.7 | 0.359 |
| Lack of coordination | 0.7 | 1.8 | 0.005 |
| Asthma | 12.4 | 15.1 | 0.047 |
| Deficiency anemias | 4.6 | 9.3 | <0.001 |
| Rheumatoid arthritis/collagen vascular diseases | 1.8 | 2.7 | 0.136 |
| Chronic pulmonary disease | 14.9 | 18.5 | 0.017 |
| Coagulopathy | 0.7 | 2.0 | 0.001 |
| Depression | 13.3 | 16.4 | 0.047 |
| Diabetes | 4.6 | 8.6 | <0.001 |
| Hypertension | 20.6 | 30.7 | <0.001 |
| Hypothyroidism | 7.0 | 7.3 | 0.800 |
| Fluid and electrolyte disorders | 4.2 | 15.1 | <0.001 |
| Obesity | 13.7 | 19.8 | <0.001 |
| Paralysis | 0.6 | 2.5 | <0.001 |
LOS, length of hospital stay.
Intraoperative Variables and Complications
Between the cohorts, there were similar proportions of duraplasty use (normal LOS, 55.3% vs. extended LOS, 59.0%; P = 0.091) (Table 3). The extended LOS cohort encountered higher rates of nerve/cord injury (normal LOS, 0.7% vs. extended LOS, 6.0%; P < 0.001) and blood transfusion (normal LOS, 0.5% vs. extended LOS, 1.4%; P = 0.020) (Table 3). Overall, significantly more patients in the extended LOS cohort encountered postoperative complications compared with the normal LOS cohort (normal LOS, 10.6% vs. extended LOS, 29.1%; P < 0.001) (Table 3). The extended LOS cohort had significantly more cerebral edema (1.2% vs. 3.5%; P < 0.001), dysphagia (1.5% vs. 5.9%; P < 0.001), acute postoperative pain (3.8% vs. 5.9%; P = 0.032), wound complication (0.1% vs. 1.8%; P < 0.001), sepsis (0.0% vs. 1.1%; P < 0.001), urinary tract infection (0.8% vs. 7.2%; P < 0.001), acute respiratory infection (1.0% vs. 6.3%; P < 0.001), deep vein thrombosis (0.0% vs. 0.8%; P < 0.001), acute posthemorrhagic anemia (1.4% vs. 2.5%; P = 0.031), and central nervous system complication (01% vs. 1.9%; P < 0.001) (Table 3). Among the patients with normal LOS, 89.4% had no postoperative complications, 9.5% had 1 complication, and 1.1% had >1 complication (Table 3). Among the patients with extended LOS, 70.9% had no postoperative complications,18.7% had no postoperative complications, and 10.4% had >1 complication (P < 0.001) (Table 3).
Table 3.
Intraoperative Variables and Postoperative Complications
| Variables (%) | Normal LOS (≤4 Days) (n = 23,159) | Extended LOS (>4 Days) (n = 6802) | P |
|---|---|---|---|
| Intraoperative | |||
| Duraplasty | 55.3 | 59.0 | 0.091 |
| Nerve/cord injury | 0.7 | 6.0 | <0.001 |
| Blood transfusion | 0.5 | 1.4 | 0.020 |
| Postoperative complications | |||
| Cerebral edema | 1.2 | 3.5 | <0.001 |
| Dysphagia | 1.5 | 5.9 | <0.001 |
| Acute postoperative pain | 3.8 | 5.9 | 0.032 |
| Wound complication | 0.1 | 1.8 | <0.001 |
| Sepsis | 0.0 | 1.1 | <0.001 |
| Renal complication | 0.1 | 0.2 | 0.434 |
| Urinary tract infection | 0.8 | 7.2 | <0.001 |
| Acute respiratory complication | 1.0 | 6.3 | <0.001 |
| Cardiac complications | 0.2 | 0.5 | 0.172 |
| Genitourinary complications | 0.1 | 0.2 | 0.434 |
| Deep vein thrombosis | 0.0 | 0.8 | <0.001 |
| Acute posthemorrhagic anemia | 1.4 | 2.5 | 0.031 |
| Central nervous system complication | 0.1 | 1.9 | <0.001 |
| Gastrointestinal complication | 0.1 | 0.4 | 0.111 |
| Any complication | 10.6 | 29.1 | <0.001 |
| Number of complications | <0.001 | ||
| 0 | 89.4 | 70.9 | |
| 1 | 9.5 | 18.7 | |
| >1 | 1.1 | 10.4 | |
LOS, length of hospital stay.
Total Cost, Discharge Disposition, and Predictors of Extended LOS
The mean LOS for the extended LOS cohort was more than twice that of the normal LOS cohort (normal LOS, 2.8 ± 0.9 days vs. extended LOS, 8.1 ± 6.4 days; P < 0.001) (Table 4). Furthermore, on average, the extended LOS cohort incurred nearly $10,000 more in total cost compared with the normal LOS cohort (normal LOS, $14,959 ± $6037 vs. extended LOS, $25,324 ± $21,629; P < 0.001) (Table 4). Compared with the normal LOS cohort, the extended LOS cohort had more patients experience nonroutine discharges (normal LOS, 7.6% vs. extended LOS, 27.0%; P < 0.001) (Table 4).
Table 4.
Postoperative Inpatient Outcomes
| Variables | Normal LOS (≤4 Days) (n = 23,159) | Extended LOS (>4 Days) (n = 6802) | P Value |
|---|---|---|---|
| Length of stay (days) | |||
| Mean ± SD | 2.8 ± 0.9 | 8.1 ± 6.4 | <0.001 |
| Median (IQR) | 3 (2—3) | 6 (5—8) | <0.001 |
| Total cost of admission ($) | |||
| Mean ± SD | 14,959 ± 6037 | 25,324 ± 21,629 | <0.001 |
| Median (IQR) | 13,866 (10,629—18,222) | 19,985 (15,183—26,940) | <0.001 |
| Disposition (%) | <0.001 | ||
| Routine | 92.3 | 72.0 | |
| Nonroutine | 7.6 | 27.0 | |
| Other | 0.0 | 1.0 | |
LOS, length of hospital stay; SD, standard deviation; IQR, interquartile range.
Stepwise multivariate regression identified black race (OR,1.55; 95% CI, 1.20–2.02), 26th-50th median household income quartile (OR, 0.72; 95% CI, 0.56–0.93), 76the100th median household income quartile (OR, 0.68; 95% CI, 0.51–0.90), private insurance (OR, 0.65; 95% CI, 0.48–0.89), obstructive hydrocephalus (OR, 3.88; 95% CI, 2.26–6.67), lack of coordination (OR, 2.12; 95% CI, 1.02–4.39), fluid and electrolyte disorders (OR, 2.54; 95% CI, 1.84–3.49), and paralysis (OR,3.35; 95% CI, 1.65–6.81) as independent risk factors associated with extended LOS (Table 5). Compared with no complications, the OR for extended LOS was 2.07 (95% CI, 1.59–2.71) for patients with 1 complication and 9.47 (95% CI, 5.86–15.30) for patients with >1 complication (Table 5).
Table 5.
Logistic Multivariate Regression Analysis on Extended Length of Stay
| Univariate Model, OR (95% CI) | Multivariate Model, OR (95% CI) | P | |
|---|---|---|---|
| Age | 1.01 (1.01—1.02) | Removed | |
| Race | |||
| White | Reference | ||
| Black | 1.87 (1.47—2.36) | 1.55 (1.20—2.02) | <0.001 |
| Hispanic | 1.68 (1.27—2.23) | 1.34 (0.99—1.82) | 0.062 |
| Other | 1.22 (0.83—1.79) | 1.11 (0.73—1.69) | 0.618 |
| Median household income quartile | |||
| 0—25th | Reference | ||
| 26th—50th | 0.68 (0.54—0.85) | 0.72 (0.56—0.93) | 0.011 |
| 51st—75th | 0.74 (0.60—0.91) | 0.80 (0.63—1.02) | 0.074 |
| 76th—100th | 0.63 (0.49—0.81) | 0.68 (0.51—0.90) | 0.007 |
| Health care coverage | |||
| Medicare | Reference | ||
| Medicaid | 0.74 (0.53—1.03) | 0.74 (0.52—1.05) | 0.094 |
| Private insurance | 0.54 (0.40—0.72) | 0.65 (0.48—0.89) | 0.007 |
| Other | 0.77 (0.52—1.15) | 0.83 (0.56—1.25) | 0.376 |
| Comorbidity | |||
| Syringomyelia | 1.45 (1.17—1.78) | 1.23 (0.95—1.59) | 0.116 |
| Obstructive hydrocephalus | 5.73 (3.40—9.67) | 3.88 (2.26—6.67) | <0.001 |
| Lack of coordination | 2.57 (1.29—5.11) | 2.12 (1.02—4.39) | 0.044 |
| Asthma | 1.26 (1.00—1.58) | Removed | |
| Deficiency anemias | 2.13 (1.54—2.96) | 1.46 (0.99—2.16) | 0.060 |
| Chronic pulmonary disease | 1.29 (1.05—1.60) | Removed | |
| Coagulopathy | 2.80 (1.46—5.35) | Removed | |
| Depression | 1.27 (1.00—1.61) | Removed | |
| Diabetes | 1.96 (1.42—2.70) | Removed | |
| Hypertension | 1.70 (1.41—2.05) | Removed | |
| Fluid and electrolyte disorders | 4.08 (3.10—5.38) | 2.54 (1.84—3.49) | <0.001 |
| Obesity | 1.55 (1.26—1.92) | Removed | |
| Paralysis | 4.26 (2.17—8.37) | 3.35 (1.65—6.81) | <0.001 |
| Blood transfusion | 2.49 (1.12—5.53) | Removed | |
| Duraplasty | 1.16 (0.98—1.38) | 1.16 (0.96—1.41) | 0.132 |
| Complications | |||
| No | Reference | ||
| 1 | 2.48 (1.90—3.23) | 2.07 (1.59—2.71) | <0.001 |
| >1 | 12.26 (7.81—19.24) | 9.47 (5.86—15.30) | <0.001 |
Bold indicates statistical significance within the univariate and the multivariate analysis. OR, odds ratio; CI, confidence interval.
DISCUSSION
In this national retrospective cohort study of 29,961 adult patients (≥18 years old) undergoing surgical decompression for CM-I, we found multiple patient-level factors that were significantly associated with extended LOS, including black race, presentation of obstructive hydrocephalus, lack of coordination, fluid and electrolyte disorders, paralysis, and inpatient postoperative complications. In addition, patients with extended LOS incurred nearly $10,000 more in total cost and had 3 times the rate of nonroutine discharges.
Identifying presenting symptoms and comorbidities in patients may not only help identify surgically indicated patients but may also help predict symptom improvement after surgery. Although there have been few data about the impact of presenting symptoms on LOS, previous literature has attempted to explore such effects on outcomes. In a retrospective study of 105 adult and pediatric patients undergoing posterior fossa decompression for CM-I, de Vlieger et al.13 showed that there was no difference in hospital stay between patients who presented and patients who did not present with syringomyelia. Similarly, in a prospective study of 61 patients undergoing foramen magnum decompression for CM-I, Vakharia et al.14 found that headache and neck pain were improved in patients with and without syrinxes, but arm pain, paresthesias, and hand tingling improved postsurgically only in the syrinx group. Analogously, in a literature review of 145 operative series of adults and children who underwent surgery for CM-I, Arnautovic et al.15 showed that the preoperative incidence of syringomyelia was 65%, with 78% corrected after surgery and 81% with resolved headaches. Furthermore, in a retrospective study of the California, Florida, and New York State Inpatient Databases, Greenberg et al.16 showed that for 1947 patients undergoing decompression for CM-I, hydrocephalus, obesity, and psychoses were correlated with an increased odds of developing surgical complications, whereas hydrocephalus, obesity, chronic lung disease, deficiency anemia, diabetes, and electrolyte disorders were associated with medical complications. Similarly, in a retrospective study of 167 adult and pediatric patients undergoing CM-I surgery, Hekman et al.17 found that the presence of syringomyelia correlated with a better Chicago Chiari Outcome Scale score after surgery, whereas sensory deficits and peripheral neuropathy correlated with a lower score. Analogous to these studies, our study found that obstructive hydrocephalus, lack of coordination, fluid and electrolyte disorders, and paralysis of some sort independently predicted prolonged LOS in patients undergoing surgery for CM-I.
Although presenting symptoms may influence longer inpatient LOS, patient demographics and hospital characteristics have also been shown to be associated with increased LOS. In their study, Greenberg et al.16 showed that patients who developed medical complications were more likely to have a longer LOS, with these patients having higher risk of being older, male, or insured by Medicare. Analogously, in their retrospective institutional study of 167 patients with CM-I, Hekman et al.17 found that younger age correlated with better outcomes on the Chicago Chiari Outcome Scale score. Although studies have previously attempted to define the impact of race on pediatric CM-I treatment, few have comparatively examined the influence of race on presentation and clinical outcome in adult patients with CM-I. In a retrospective study of 936 pediatric CM-I surgeries using statewide inpatient databases for California, Florida, and New York, Greenberg et al.18 showed that black patients were 3.9 times more likely to encounter postoperative medical complications compared with their white patient counterparts. Similarly, in a study of 287 adult patients with CM-I, Krucoff et al.19 showed that whereas white patients presented with a greater frequency of back pain, ataxia, and syncope, black patients had worse lower extremity weakness, tonsillar ectopy, and syringomyelia. Furthermore, black patients trended to have a mean LOS 1 day longer compared with white patients.19 These investigators also suggested that the increased LOS in the public insurance cohort was caused by the trend of increasing age.19 Our study adds to these findings, suggesting that black race increases the likelihood of encountering extended LOS, whereas private insurance and higher income quartile predict a reduced likelihood. Further studies are necessary to identify the impact that patient demographics may have on inpatient hospital LOS.
As variability in U.S. hospital patterns and treatment philosophy continues to increase, some studies have started to turn their attention to examining duraplasty as a significant predictor for extended LOS. For example, in a retrospective institutional study of 42 patients undergoing decompression for CM-I, Chotai et al.20 showed that patients who had a duraplasty had a 2.7 times greater likelihood of longer LOS as well as an increased risk for complications compared with those who did not. Similarly, in a retrospective cohort study of 57 institutional patients undergoing posterior fossa decompression with or without duraplasty, Li et al.21 showed that not having the duraplasty resulted in shorter LOS and fewer cerebrospinal fluid–related complications. Other studies have found no difference in LOS with treatment choice. In a retrospective study of 49 patients undergoing posterior fossa decompression with either a dura-splitting or combined technique (duraplasty with arachnoid dissection and coagulation of the herniated tonsils) for CM-I, Geng et al.22 showed that there was no difference in LOS; however, operating times were higher with the combined technique. Similar to that study, we found that choice of duraplasty was not associated with extended LOS.
In an era of increased importance placed on identifying at-risk patients in the inpatient setting, it is necessary to determine the impact of inpatient surgical factors that may affect LOS and health care costs. Although multiple pediatric studies have attempted to identify inpatient complications as a predictor for LOS, few have looked at the impact in adult CM-I decompression. In their study of 1947 treated adults with CM-I, Greenberg et al.16 found that the most common surgical complications that were neurosurgically specific were cerebrospinal fluid leak and pseudomeningocele (13.5%), shunt-related complications (4.9%), meningitis (4.8%), and wound infection (3.2%). In addition, the most common medical complications were a pulmonary complication/pneumonia (2.6%) and urinary–renal complications (0.98%).16 These investigators found that patients with a surgical or medical complication were independently associated with increased length of hospital stay, with estimated LOS being 10.5 days (95% CI, 9.2–11.8 days) and 13.9 days (95% CI, 11.8–16.0 days), respectively, compared with no complication, which averaged 4.1 days (95% CI, 4.1–4.2 days).16 In addition, the costs associated with patients experiencing surgical and medical complications were $32,507 (95% CI, $29,516–$35,497) and $51,100 (95% CI, $46,659–$55,543), compared with no complication, which averaged $17,832 (95% CI, $17,707–$17,958) for indexed hospitalization. Analogous to that study, our study found that having ≥1 complication significantly increased the odds of encountering an extended LOS. In addition, our study showed a total cost of $14,959 for those with normal LOS and a total cost of $25,324 for those with extended LOS. Identifying the specific complications influencing LOS may provide avenues for quality care initiatives to optimize care and reduce health care costs.
This study has several inherent limitations common to administrative databases, which have potential implications for its interpretation. First, the analysis is retrospective, with data available only for ICD-9-CM codes, which may contain coding and reporting biases. Second, there is a possibility of misclassified or incomplete data. We are also unable to comment on the severity and range of potentially important clinical factors such as preoperative motor and sensory functional status and central nervous system complication encountered. Furthermore, this study is limited by lack of data on surgeon preference, unique perioperational and surgical-related risks, patient care outside the United States, and grouped intercohort differences among patients with extended LOS. Because the NIS has information specific to only 1 inpatient admission, we are unable to comment on differences in long-term functional outcomes. Regardless, this study sheds light on important patient-level and hospital-level factors associated with extended LOS for patients with CM-I undergoing decompression surgery.
CONCLUSIONS
Our study shows that extended LOS after adult CM-I decompression surgery may be influenced by multiple modifiable and nonmodifiable patient demographics, baseline presenting symptoms and comorbidities, and postoperative complications. Factors identified in our study include black race, presentation of obstructive hydrocephalus, lack of coordination, fluid and electrolyte disorders, paralysis, and inpatient postoperative complications. In addition, patients with extended LOS incurred nearly $10,000 more in total cost and had 3 times the rate of nonroutine discharges.
Abbreviations and Acronyms
- CI
Confidence interval
- CM-I
Chiari malformation type I
- ICD-9-CM
International Classification of Diseases: Ninth Revision: Clinical Modification
- LOS
Length of hospital stay
- NIS
National Inpatient Sample
- OR
Odds ratio
Footnotes
Conflict of interest statement:
The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
REFERENCES
- 1.Rossi VJ, Ahn J, Bohl DD, Tabaraee E, Singh K. Economic factors in the future delivery of spinal healthcare. World J Orthop. 2015;6:409–412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Aboulezz AO, Sartor K, Geyer CA, Gado MH. Position of cerebellar tonsils in the normal population and in patients with Chiari malformation: a quantitative approach with MR imaging. J Comput Assist Tomogr. 1985;9:1033–1036. [DOI] [PubMed] [Google Scholar]
- 3.Krell RW, Girotti ME, Dimick JB. Extended length of stay after surgery: complications, inefficient practice, or sick patients? JAMA Surg. 2014;149: 815–820. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cohen ME, Bilimoria KY, Ko CY, Richards K, Hall BL. Variability in length of stay after colorectal surgery: assessment of 182 hospitals in the national surgical quality improvement program. Ann Surg. 2009;250:901–907. [DOI] [PubMed] [Google Scholar]
- 5.Ansari SF, Yan H, Zou J, Worth RM, Barbaro NM. Hospital length of stay and readmission rate for neurosurgical patients. Neurosurgery. 2018;82: 173–181. [DOI] [PubMed] [Google Scholar]
- 6.Lee SY, Lee SH, Tan JHH, et al. Factors associated with prolonged length of stay for elective hepatobiliary and neurosurgery patients: a retrospective medical record review. BMC Health Serv Res. 2018;18:5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Barkovich AJ, Wippold FJ, Sherman JL, Citrin CM. Significance of cerebellar tonsillar position on MR. AJNR Am J Neuroradiol. 1986;7:795–800. [PMC free article] [PubMed] [Google Scholar]
- 8.Rekate HL. Natural history of the Chiari type I anomaly. J Neurosurg Pediatr. 2008;2:177–178. [DOI] [PubMed] [Google Scholar]
- 9.Fernández AA, Guerrero AI, Martínez MI, et al. Malformations of the craniocervical junction (Chiari type I and syringomyelia: classification, diagnosis and treatment). BMC Musculoskelet Disord. 2009;10(suppl 1):S1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fischbein R, Saling JR, Marty P, et al. Patient-reported Chiari malformation type I symptoms and diagnostic experiences: a report from the national Conquer Chiari Patient Registry database. Neurol Sci. 2015;36:1617–1624. [DOI] [PubMed] [Google Scholar]
- 11.Wilkinson DA, Johnson K, Garton HJ, Muraszko KM, Maher CO. Trends in surgical treatment of Chiari malformation Type I in the United States. J Neurosurg Pediatr. 2017;19:208–216. [DOI] [PubMed] [Google Scholar]
- 12.Ladner TR, Greenberg JK, Guerrero N, et al. Chiari malformation Type I surgery in pediatric patients. Part 1: validation of an ICD-9-CM code search algorithm. J Neurosurg Pediatr. 2016;17: 519–524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.De Vlieger J, Dejaegher J, Van Calenbergh F. Posterior fossa decompression for Chiari malformation type I: clinical and radiological presentation, outcome and complications in a retrospective series of 105 procedures. Acta Neurol Belg. 2019;119:245–252. [DOI] [PubMed] [Google Scholar]
- 14.Vakharia VN, Guilfoyle MR, Laing RJ. Prospective study of outcome of foramen magnum decompressions in patients with syrinx and nonsyrinx associated Chiari malformations. Br J Neurosurg. 2012;26:7–11. [DOI] [PubMed] [Google Scholar]
- 15.Arnautovic A, Splavski B, Boop FA, Arnautovic KI. Pediatric and adult Chiari malformation type I surgical series 1965–2013: a review of demographics, operative treatment, and outcomes. J Neurosurg Pediatr. 2015;15:161–177. [DOI] [PubMed] [Google Scholar]
- 16.Greenberg JK, Ladner TR, Olsen MA, et al. Complications and resource use associated with surgery for Chiari malformation type 1 in adults: a population perspective. Neurosurgery. 2015;77: 261–268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hekman KE, Aliaga L, Straus D, et al. Positive and negative predictors for good outcome after decompressive surgery for Chiari malformation type 1 as scored on the Chicago Chiari Outcome Scale. Neurol Res. 2012;34:694–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Greenberg JK, Olsen MA, Yarbrough CK, et al. Chiari malformation type I surgery in pediatric patients. Part 2: complications and the influence of comorbid disease in California, Florida, and New York. J Neurosurg Pediatr. 2016;17:525–532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Krucoff MO, Cook S, Adogwa O, et al. Racial, socioeconomic, and gender disparities in the presentation, treatment, and outcomes of adult Chiari I malformations. World Neurosurg. 2017;97: 431–437. [DOI] [PubMed] [Google Scholar]
- 20.Chotai S, Medhkour A. Surgical outcomes after posterior fossa decompression with and without duraplasty in Chiari malformation-I. Clin Neurol Neurosurg. 2014;125:182–188. [DOI] [PubMed] [Google Scholar]
- 21.Li HY, Li YM, Chen H, Li Y, Shi XW. [Comparison of posterior fossa decompression with and without duraplasty for surgical management for adult Chiari malformation type]. Zhonghua Yi Xue Za Zhi. 2017;97:1947–1950 [in Chinese]. [DOI] [PubMed] [Google Scholar]
- 22.Geng LY, Liu X, Zhang YS, et al. Dura-splitting versus a combined technique for Chiari malformation type I complicated with syringomyelia. Br J Neurosurg. 2018. 10.1080/02688697.2018.1498448. [DOI] [PubMed] [Google Scholar]