Publish in this journal
Journal Information
Vol. 28. Num. 1.January - February 2017Pages 1-50
Vol. 28. Num. 1.January - February 2017Pages 1-50
Clinical Research
DOI: 10.1016/j.neucie.2016.06.002
Incidence and risk factors of 30-day readmission in neurosurgical patients
Incidencia y factores de riesgo de reingreso hospitalario a los 30 días en pacientes neuroquirúrgicos
Antonio José Vargas López
Corresponding author

Corresponding author.
, Carlos Fernández Carballal
Servicio de Neurocirugía, Hospital General Universitario Gregorio Marañón, Madrid, Spain
Article information
Full Text
Download PDF
Figures (1)
Tables (5)
Table 1. Summary of patients’ demographic data and type of procedure conducted.
Table 2. Case study and frequency of readmission of the main diagnostic groups, as well as the cause of readmission and its frequency.
Table 3. Different diagnoses that led to readmission with their corresponding frequencies.
Table 4. Frequency of readmissions in other specialities described in previous work.
Table 5. Frequency of readmissions in neurosurgery described in previous works.
Show moreShow less

The 30-day readmission rate has become an important indicator of health care quality. This study focuses on the incidence of 30-day readmission in neurosurgical patients and related risk factors.

Material and methods

A retrospective review was performed on patients treated in a neurosurgery department between 1 January 2012 and the 31 December 2013. Patients requiring readmission within 30 days of discharge and the readmission diagnosis were identified, and the factors related to their readmission were analysed.


A total of 1854 interventions were carried out on 1739 patients during the aforementioned (study) period. Of the remaining patients, 174 (10.2%) required readmission within 30 days of discharge. The main causes of readmission were problems related to the surgical wound (21.2% of all readmissions), followed by respiratory processes (18.8%). A total of 73.9% of readmissions occurred in patients who had undergone cranial surgery. Multiple comorbidities estimated by Charlson comorbidity index and length of hospital stay were identified as factors related to a higher readmission rate.


The 30-day readmission rate observed in our series was 10.2%. Multiple comorbidity expressed by the Charlson comorbidity index and length of hospital stay were related to readmission.

Health care cost
Quality improvement

El reingreso a los 30días se ha convertido en un parámetro de uso creciente como indicador de calidad asistencial. El presente trabajo pretende establecer la frecuencia de reingreso a los 30días entre pacientes que precisaron intervención neuroquirúrgica, así como analizar los factores relacionados con dicha eventualidad.

Material y métodos

Se han revisado de forma retrospectiva los pacientes intervenidos en nuestra institución desde el 1 de enero de 2012 hasta el 31 de diciembre de 2013. Se han identificado los pacientes que precisaron reingreso en los primeros 30días tras recibir el alta hospitalaria, así como la causa que motivó dicho ingreso. Se han analizado los factores relacionados con el reingreso.


Se llevaron a cabo 1.854 intervenciones en 1.739 pacientes durante el período señalado. Durante el ingreso fallecieron 36 pacientes (2,1%). De los pacientes restantes, un total de 174 (10,2%) precisaron reingreso hospitalario en los primeros 30días tras el alta. La principal causa de reingreso estuvo representada por los problemas relacionados con la herida quirúrgica (21,2% del total de reingresos), seguida de los procesos respiratorios (18,8%). El 73,9% de los reingresos aconteció en pacientes en los que se había realizado cirugía craneal. Los factores relacionados con una mayor tasa de reingreso fueron la comorbilidad múltiple estimada por el índice de comorbilidad de Charlson y la duración de la estancia hospitalaria anterior al reingreso.


En nuestra serie el 10,2% de los pacientes precisaron nuevo ingreso hospitalario a los 30días. La comorbilidad múltiple expresada por el índice de comorbilidad de Charlson y la duración de la estancia hospitalaria estuvieron relacionados con dicha eventualidad.

Palabras clave:
Coste sanitario
Mejoría en calidad
Full Text

Evaluation of heath care quality has become imperative in the last few years. One of the most widely used indicators of quality is 30-day hospital readmission.1 While on occasions this result cannot be completely avoided, it is considered to represent a reflection of the efficacy of the treatment given, as well as of the morbidity associated with said treatment.

Early hospital readmission, in addition to causing obvious negative consequences in the quality of life of the patient and a potential increase in the risk of nosocomial infection,2 also has significant financial impact. The cost of readmission in the Medicare population in one year is estimated to top 17 billion dollars.3

Readmission frequency, as well as associated factors, can vary among the various specialties.4,5 In fact, the causes that make it worse are different between patients that have received only medical care versus those that have received surgical treatment.6,7

The present work, therefore, arises from the need to know both the frequency of 30 day hospital readmission, as well as the factors associated to it in neurosurgical patients.


Patients that needed neurosurgical intervention at Hospital General Universitario Gregorio Marañón were retrospectively analysed for the period starting on 1 January 2012 until 31 December 2013. The procedures were collected from the surgery record belonging to the Neurosurgery Department. Endovascular procedures, intracranial pressure monitoring of neurocritical patients, external ventricular drains and outpatient procedures have been excluded.

Data belonging to the diagnosis that led to initial admission (also called index admission) and the treatment given were obtained. According to the index diagnosis, the following 6 diagnosis groups were created: patients with cerebral tumours, hydrocephalus, shunt malfunction, chronic subdural haematomas, spine pathology that needs instrumentation and non-instrumented spine pathology. Demographic variables (age and gender), medical-surgical history and the duration of the hospital stay during the initial admission were analysed in each of the patients. The subgroup of paediatric patients includes those that are younger than 18 years of age. The impact of the medical comorbidities of each patient has been estimated using the Charlson comorbidity index (Fig. 1). The information for these variables was obtained from the medical histories of the patients.

Fig. 1.

Charlson comorbidity index.

The patients that needed to be readmitted to any of the departments at our centre during the first 30 days after being discharged from the hospital were identified, with the corresponding diagnosis that led to readmission. Scheduled readmissions for procedures or administration of chemotherapy treatment, as well as those admitted for supplemental testing were excluded.

Univariate analysis to identify the impact of each of the readmission variables separately has been carried out using the chi-squared and Student's t parametric tests. In addition, multivariate analysis has been performed using the logistic regression model. The variables of age, separating patients older than 65 years from the rest, and the initial diagnosis were dichotomised, differentiating patients with cranial pathology from patients with spinal pathology or another different diagnosis. SPSS 15.0 software was used. Statistical significance was established at p<0.05.


During the period specified, 1854 procedures were performed on 1739 patients. A total of 86 patients (4.9%) needed more than one procedure during the index admission. Of all the patients, 932 were men (53.6%) and 807 were women (46.4%). The average age was 54.19±19.5years (0–88). In total, 1770 procedures (95.5%) were performed on adult patients, while 84 procedures were performed on paediatric patients (4.5%). The comorbidities calculation using the Charlson comorbidity index produced an average of 2.61±1.9 points (0–9) (Table 1). Of all the patients, 239 (12.9%) presented a spreading oncological process prior to initial admission.

Table 1.

Summary of patients’ demographic data and type of procedure conducted.

Males  932 (53.6%) 
Females  807 (46.4% 
Age  54.19±19.5 years (0–88) 
Adult patients  1770 (95.5%) 
Paediatric patients (<18 years)  84 (4.5%) 
Charlson comorbidity index  2.61±1.9 points (0–9) 
Scheduled surgery  1378 (74.3%) 
Emergency surgery  476 (25.7%) 
Cranial surgery  1183 (63.8%) 
Spinal/other surgeries  671 (36.2%) 

1378 scheduled procedures were carried out, while 476 were urgent. Of all the procedures, 1183 were cranial surgeries and 671 were spinal or of another type. The average hospital stay during the initial admission was 8 days (1–148). During that period, 36 (2.1%) patients died.

A total of 174 patients (10.2%) needed to be readmitted in the first 30 days after discharge. Table 2 details the data on the frequency of readmission and the causes of readmission for each diagnostic group. 66.1% of readmissions (115 patients) were patients with a history of cranial surgery, while 59 (33.9%) patients readmitted had received spinal surgery or some other type of operation.

Table 2.

Case study and frequency of readmission of the main diagnostic groups, as well as the cause of readmission and its frequency.

Brain tumours 
Total cases: 329 
Readmissions: 55 (16.7%) 
Readmission diagnosis 
Infection/fistula: 18 cases 
Pneumonia/RF: 8 cases 
UTI: 6 cases 
Shunt malfunction: 1 case 
DVT: 11 cases 
Epileptic crisis: 2 cases 
Others: 9 cases 
Total cases: 111 
Readmissions: 5 (4.5%) 
Readmission diagnosis 
Infection/fistula: 1 case 
Pneumonia/RF: 0 cases 
UTI: 0 cases 
Shunt malfunction: 4 cases 
DVT: 0 cases 
Epileptic crisis: 0 cases 
Others: 0 cases 
CSF shunt malfunction 
Total cases: 84 
Readmissions: 13 (15.5%) 
Readmission diagnosis 
Infection/fistula: 1 case 
Pneumonia/RF: 0 cases 
UTI: 0 cases 
Shunt malfunction: 12 cases 
DVT: 0 cases 
Epileptic crisis: 0 cases 
Others: 0 cases 
Chronic subdural haematoma 
Total cases: 92 
Readmissions: 16 (17.4%) 
Readmission diagnosis 
Infection/fistula: 1 case 
Pneumonia/RF: 5 cases 
UTI: 3 cases 
Shunt malfunction: 0 cases 
DVT: 1 case 
Epileptic crisis: 1 case 
Others: 5 cases 
Non-instrumented spinal pathology 
Total cases: 371 
Readmissions: 11 (3.0%) 
Readmission diagnosis 
Infection/fistula: 5 cases 
Pneumonia/RF: 1 case 
UTI: 1 case 
Shunt malfunction: 0 cases 
DVT: 1 case 
Epileptic crisis: 0 cases 
Others: 3 cases 
Instrumented spinal pathology 
Total cases: 322 
Readmissions: 21 (6.5%) 
Readmission diagnosis 
Infection/fistula: 7 cases 
Pneumonia/RF: 3 cases 
UTI: 4 cases 
Shunt malfunction: 0 cases 
DVT: 2 cases 
Epileptic crisis: 0 cases 
Others: 5 cases 

The causes for readmission are detailed in Table 3. The main cause was infection due to a surgical wound and/or cerebrospinal fluid (CSF) fistula, which occurred in 36 patients (21.2%), followed by pneumonia or respiratory failure in 31 cases (18.8%).

Table 3.

Different diagnoses that led to readmission with their corresponding frequencies.

Readmission diagnosis  n (%) 
Wound infection/CSF fistula  36 (21.2) 
Pneumonia/respiratory failure  31 (18.8) 
UTI  21 (14.8) 
VPS malfunction  19 (13.1) 
DVT  18 (12.3) 
Epileptic crisis  7 (4.5) 
Other  23 (15.3) 

The univariate statistical analysis showed a relationship between 30-day readmission and patients over 65 years old with a history of diabetes mellitus, spreading oncological disease prior to initial admission, a greater score on the Charlson comorbidity index and a greater duration in days of the initial admission. These last 2 parameters were identified in the multivariate analysis as factors associated to readmission, with an OR of 1.71 (CI 95%: 1.12–2.94) for the Charlson index and an OR of 1.30 (CI 95%: 1.14–1.50) for the hospital stay. No relationship was found between readmission and the variables of gender, surgery in paediatric or adult patients, type of procedure (cranial versus spinal-others) and urgent or non-urgent character of the procedure.


Publications focused on readmission have significantly increased since 2010. The Patient Protection and Affordable Care Act (PPACA) will develop a readmission reduction programme in the United States to be implemented at the beginning of fiscal year 2013.8 Said programme has considered an increase in penalisation for the centres that exceeded certain readmission figures. Said penalisation would be from 1 to 3% of the financial allowance destined for each diagnostic group in the Medicare programme.8

The need for readmission may vary based on each medical speciality. There are publications that state the frequency of readmission for general surgery to be around 11.3%,4–10 while for patients undergoing a coronary bypass, the frequency of readmission is 16.5%.11 The work of Friedman in 2004 estimated the global rate of 30-day readmission in the Medicare population to be 19.5%12 (Table 4).

Table 4.

Frequency of readmissions in other specialities described in previous work.

Main author (year of publication)  Patient group  Readmission % 
Friedman and Basu (2014)12  Medicare population  19.5 
Kassin et al. (2012)4  General surgery  11.3 
Schneider et al. (2012)10  Colorectal surgery  9–15 
Reddy et al. (2009)9  Pancreatectomy  16 
Hannan et al. (2011)11  Coronary bypass  16.5 

Recent studies estimate the global frequency of readmission in neurosurgical patients to be between 6.8 and 11.8%.13,14 In our series that frequency was 10.2%. In the separate analysis of spinal pathology and cranial pathology, differences in frequency of readmission were seen. Thus, the patients that needed spinal procedures had to be readmitted in about 4.2–9.1% of cases,13–17 while readmission after cranial surgery was about 14–24%18 (Table 5). This study, however, has not found significant differences in frequency of readmission between the two groups.

Table 5.

Frequency of readmissions in neurosurgery described in previous works.

Main author (year of publication)  Patient group  Readmission % 
Buchanan et al. (2014)13  General neurosurgery  6.8 
Shah et al. (2013)20  General neurosurgery  11.8 
Moghavem et al. (2015)18  Cranial surgery  14–24 
Marcus et al. (2014)19  Brain tumours  13.2 
Bernatz et al. (2015)15  Rachis  4.7–6.5 

Among the diagnoses that led to readmission, the most common in our series was represented by problems related to a surgical wound in 21.2% of cases. Previous publications have put said frequency at 11–32% of readmissions in neurosurgical patients.2,16,20 The breakdown of readmission diagnoses in this study shows that at least 34.4% of readmissions (due to problems with a surgical wound or malfunction of the CSF shunt system) are directly related to the surgical procedure.

In accordance with the results obtained in our research, previous studies have shown that a longer hospital stay during the index admission is associated with a greater risk of readmission.2,3,13,15 In this case it is believed that prolonging the stay would be a surrogate variable of the existence of postoperative complications.4,10 Apart from other disadvantages, prolonging the hospital stay may cause nosocomial infections that would lead to readmission.19

Previous publications have described a direct relationship between a greater number of comorbidities present at the time of initial admission and the need for 30-day readmission. The Charlson comorbidity index was used in our study to assign a value proportional to the comorbidity score. The results obtained in our experience confirm this relationship.

In relation to the readmission frequency per diagnostic groups, it was 16.7% for brain tumours. This frequency is slightly higher than that described in the work of Marcus,19 where it was 13.2%. Within this group, the most common cause of readmission observed in our research were problems related to a surgical wound. The group made up of patients that had procedures for chronic subdural haematomas had a higher frequency of readmission in our study, at 17.4%. In this group, unlike what occurs with brain tumours, most causes of readmission were due to pathologies secondary to medical care, such as respiratory processes, urinary infections, venous thrombosis and other medical diagnoses.

A greater understanding of the factors associated with readmission could make it possible to design programmes to decrease its frequency. The creation of strategies capable of reducing readmission would result in a significant reduction in healthcare costs, especially since previous publications put readmission costs of patients previously treated for brain tumours at more than 20,000 dollars.19 Therefore, studies have been conducted which suggest that adequately educating the patient, together with careful recommendations regarding using drugs as part of the information given at discharge, would reduce 30-day readmission.21,22 Other types of actions would be based on detecting the emerging onset of postoperative complications early which would allow outpatient treatment to be established that was capable of anticipating and stopping the full development of a process that could lead to readmission.19 Therefore, follow-up of high-risk patients subsequent to discharge is necessary.23 In this regard, The Agency for Healthcare Research and Quality has developed a follow-up programme in a plan intended to reduce readmission rates after discharge. The frequency established for this outpatient follow-up in neurosurgical patients varies depending on the institution. Marcus et al.19 sets the time of this evaluation at day 11 after the end of hospitalisation, coinciding with the average number of days lapsed since discharge in patients that needed to be readmitted.

This research is, as far as we know, the first at a national level directed at identifying readmission frequency in neurosurgical patients, as well as the factors associated with it. The limitations of this study are the fact that the data is only from one institution. Since administrative data was not used, readmission from other centres could not be recorded. This has also denied obtaining a larger sample size that would lead to an increase in the statistical power of the results. On the other hand, and in this study's favour, is the fact that as the authors obtained the information directly from medical histories, said information is more reliable. This data collection method also avoids incorrect information that could occur in studies based on administrative data, due to inconsistency in data codification, the difficulty of administrative data in reflecting significant changes in the development of the disease, as well as the fact that this data is oriented for purely administrative purposes.24–27

In conclusion, understanding the frequency of readmission in neurosurgical patients and the associated constraints represents a necessary area of study in order to improve quality of care for the patient's benefit and to optimise the financial cost. Subsequent studies are needed that are oriented at characterising factors associated with readmission better and evaluating potential procedures in order to minimise readmission.


In our series, 10.2% of patients required another admission at 30 days. A third of the cases were directly related to the surgical procedure performed during the previous admission. Multiple comorbidity expressed by the Charlson comorbidity index and the duration of the hospital stay were identified as risk factors independent of a greater probability of 30-day readmission.

Conflicts of interest

The authors declare that there are no conflicts of interest.

J. Benbassat,M. Taragin
Hospital readmissions as a measure of quality of health care: advantages and limitations
Arch Intern Med, 160 (2000), pp. 1074-1081
R.P. Kiran,C.P. Delaney,A.J. Senagore,M. Steel,T. Garafalo,V.W. Fazio
Outcomes and prediction of hospital readmission after intestinal surgery
J Am Coll Surg, 198 (2004), pp. 877-883
S.F. Jencks,M.V. Williams,E.A. Coleman
Rehospitalizations among patients in the Medicare fee-for-service program
N Engl J Med, 360 (2009), pp. 1418-1428
M.T. Kassin,R.M. Owen,S.D. Perez,I. Leeds,J.C. Cox,K. Schnier
Risk factors for 30-day hospital readmission among general surgery patients
J Am Coll Surg, 215 (2012), pp. 322-330
S. Sarda,M. Bookland,J. Chu,M.M. Shoja,M.P. Miller,S.B. Reisner
Return to system within 30 days of discharge following pediatric non-shunt surgery
J Neurosurg Pediatr, 14 (2014), pp. 654-661
S.F. Khuri,W.G. Henderson,R.G. DePalma,C. Mosca,N.A. Healey,D.J. Kumbhan
Determinants of long-term survival after major surgery and the adverse effect of postoperative complications
Ann Surg, 242 (2005), pp. 326-341
discusión 341-3
G.F. Anderson,E.P. Steinberg
Hospital readmissions in the Medicare population
N Engl J Med, 311 (1984), pp. 1349-1353
K. Dharmarajan,A.F. Hsieh,Z. Lin,H. Bueno,J.S. Ross,L.I. Horwitz
Diagnoses and timing of 30-day readmissions alter hospitalization for heart failure, acute myocardial infarction, or pneumonia
D.M. Reddy,C.M. Townsend Jr.,Y.F. Kuo,J.L. Freeman,J.S. Goodwin,T.S. Riall
Readmission after pancreatectomy for pancreatic cancer in Medicare patients
J Gastrointest Surg, 13 (2009), pp. 1963-1974
discusión 1974-5
E.B. Schneider,O. Hyder,B.S. Brooke,J. Efron,J.L. Cameron,B.H. Edil
Patient readmission and mortality after colorectal surgery for colon cancer: impact of length of stay relative to other clinical factors
J Am Coll Surg, 214 (2012), pp. 390-398
discusión 398-9
E.L. Hannan,Y. Zhong,S.J. Lahey,A.T. Culliford,J.P. Gold,C.R. Smith
30-day readmissions after coronary artery bypass graft surgery in New York State
JACC Cardiovasc Interv, 4 (2011), pp. 569-576
B. Friedman,J. Basu
The rate and cost of hospital readmissions for preventable conditions
Med Care Res Rev, 61 (2004), pp. 225-240
C.C. Buchanan,E.A. Hernandez,J.M. Anderson,J.A. Dye,M. Leung,F. Buxey
Analysis of 30-day readmissions among neurosurgical patients: surgical complication avoidance as key to quality improvement
J Neurosurg, 121 (2014), pp. 170-175
M.C. Wang,M. Shivakoti,R.A. Sparapani,C. Guo,P.W. Laud,A.B. Nattinger
Thirty-day readmissions after elective spine surgery for degenerative conditions among US Medicare beneficiaries
J.T. Bernatz,P.A. Anderson
Thirty-day readmission rates in spine surgery: systematic review and meta-analysis
Neurosurg Focus, 39 (2015), pp. E7
R.A. McCormack,T. Hunter,N. Ramos,R. Michels,L. Hutzler,J.A. Bosco
An analysis of causes of readmission after spine surgery
B.D. Kim,T.R. Smith,S. Lim,G.R. Cybulski,J.Y. Kim
Predictors of unplanned readmission in patients undergoing lumbar decompression: multi-institutional analysis of 7016 patients
J Neurosurg Spine, 20 (2014), pp. 606-616
N. Moghavem,D. Morrison,J.K. Ratliff,T. Hernandez-Boussard
Cranial neurosurgical 30-day readmissions by clinical indication
J Neurosurg, 123 (2015), pp. 189-197
L.P. Marcus,B.A. McCutcheon,A. Noorbakhsh,R.P. Parina,D.D. Gonda,C. Chen
Incidence and predictors of 30-day readmission for patients discharged home after craniotomy for malignant supratentorial tumors in California (1995–2010)
J Neurosurg, 120 (2014), pp. 1201-1211
M.N. Shah,I.T. Stoev,D.E. Sanford,F. Gao,P. Santiago,D.P. Jaques
Are readmission rates on a neurosurgical service indicators of quality of care
J Neurosurg, 119 (2013), pp. 1043-1049
B.W. Jack,V.K. Chetty,D. Anthony,J.L. Greenwald,G.M. Sanchez,A.E. Johnson
A reengineered hospital discharge program to decrease rehospitalization: a randomized trial
Ann Intern Med, 150 (2009), pp. 178-187
B.E. Koehler,K.M. Richter,L. Youngblood,B.A. Cohen,I.D. Prengler,D. Cheng
Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle
J Hosp Med, 4 (2009), pp. 211-218
L.O. Hansen,R.S. Young,K. Hinami,A. Leung,M.V. Williams
Interventions to reduce 30-day rehospitalization: a systematic review
P.J. Holt,J.D. Poloniecki,M.M. Thompson
Multicentre study of the quality of a large administrative data set and implications for comparing death rates
Br J Surg, 99 (2012), pp. 58-65
P.S. Keenan,S.L. Normand,Z. Lin,E.E. Drye,K.R. Bhat,J.S. Ross
An administrative claims measure suitable for profiling hospital performance on the basis of 30-day all-cause readmission rates among patients with heart failure
Circ Cardiovasc Qual Outcomes, 1 (2008), pp. 29-37
H.M. Krumholz,Z. Lin,E.E. Drye,M.M. Desai,L.F. Han,M.T. Rapp
An administrative claims measure suitable for profiling hospital performance based on 30-day all-cause readmission rates among patients with acute myocardial infarction
Circ Cardiovasc Qual Outcomes, 4 (2011), pp. 243-252
L.I. Iezzoni
Assessing quality using administrative data
Ann Intern Med, 127 (1997), pp. 666-674

Please cite this article as: Vargas López AJ, Fernández Carballal C. Incidencia y factores de riesgo de reingreso hospitalario a los 30 días en pacientes neuroquirúrgicos. Neurocirugia. 2017;28:22–27.

Copyright © 2017. Sociedad Española de Neurocirugía
Neurocirugía (English edition)

Subscribe to our Newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.