ORIGINAL ARTICLE
Temporal trends in head and neck cancer surgery reconstruction Zhen Gooi, MBBS,1 Christine G. Gourin, MD, MPH,1,2* Kofi D. O. Boahene, MD,1 Patrick J. Byrne, MD,1 Jeremy D. Richmon, MD1 1
Department of Otolaryngology–Head and Neck Surgery, Johns Hopkins University, Baltimore, Maryland, 2Armstrong Institute for Patient Safety and Quality, Johns Hopkins University, Baltimore, Maryland.
Accepted 28 May 2014 Published online 21 July 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/hed.23786
ABSTRACT: Background. The purpose of this study was to analyze changing trends in head and neck cancer reconstructive surgery and analyze the effect of surgeon and hospital volume. Methods. Data from the Nationwide Inpatient Sample (NIS) for 133,850 patients who underwent a major ablative procedure for a malignant oral cavity, laryngeal, hypopharyngeal, or oropharyngeal neoplasm in 1993 to 2010 were analyzed using cross-tabulations and multivariate regression. Results. Reconstructive surgery in 2001 to 2010 was significantly associated with prior radiation (odds ratio [OR] 5 2.6; 95% confidence interval [95% CI] 5 1.4–4.9), comorbidity (OR 5 1.6; 95% CI 5 1.1–2.2), laryngeal cancer (OR 5 0.7; 95% CI 5 0.6–0.9), oropharyngeal cancer (OR 5 0.5; 95% CI 5 0.4–0.7), high-volume hospitals (OR 5 3.9; 95% CI 5 1.5–10.2), and high-volume surgeons (OR 5 2.0; 95% CI 5 1.1–
INTRODUCTION The oncologic resection of head and neck cancer incurs functional consequences to patients affecting speech, swallowing, and appearance. In efforts to restore normal physiology after surgical resection, a variety of techniques are used by reconstructive surgeons using pedicled or free tissue flaps. The safety and efficacy of these reconstructive techniques have been demonstrated in multiple large case series with low failure rates.1–5 Over the past 2 decades, there has been a paradigm shift in the primary treatment of laryngeal and oropharyngeal cancers, with increased utilization of chemoradiation compared to surgery.6–8 Consequently, the rate of salvage surgery for both laryngeal and oropharyngeal cancers nationally has been shown to be increasing, with a gradual shift in the performance of surgery to high volume centers and by high volume surgeons.9,10 It is currently unknown if the use of flap reconstruction in the era of increasing salvage surgery mirrors these patterns. Reconstruction of head and neck defects after oncologic resection, particularly those involving free tissue transfers, are complex procedures requiring advanced surgical training and close postoperative monitoring. The relationship between increased hospital volume, surgeon volume, and
*Corresponding author: C. G. Gourin, Johns Hopkins Outpatient Center, Department of Otolaryngology–Head and Neck Surgery, 601 N. Caroline Street, Suite 6260, Baltimore, MD 21287. E-mail: [email protected]
3.9), compared to 1993–2000. Reconstruction by high-volume surgeons was significantly associated with prior radiation (OR 5 1.8; 95% CI 5 1.1–3.1) and lower in-hospital mortality (OR 5 0.3; 95% CI 5 0.1– 1.0). A statistically significant negative correlation was observed between high-volume surgeons and length of hospitalization and hospital-related costs. Conclusion. These data reflect changing trends in head and neck cancer reconstructive surgery, with meaningful differences in the type of surgiC 2014 Wiley Periodicals, cal care provided by high-volume surgeons. V Inc. Head Neck 37: 1509–1517, 2015
KEY WORDS: hospital-acquired conditions, complications, surgery, head and neck neoplasms, costs, Nationwide Inpatient Sample
improved postoperative outcomes for complicated intrathoracic and intra-abdominal procedures measured through indices of short-term and long-term mortality have been demonstrated in a number of studies.11–13 We undertook this study to evaluate temporal trends in head and neck cancer reconstruction and variables associated with in-hospital mortality, postoperative complications, length of stay, and costs using a national hospital discharge database. We hypothesized that there is a trend toward centralization of care by higher volume surgeons and that this trend is associated with improved postoperative outcomes, similar to improved outcomes demonstrated for high-volume care for intrathoracic and intraabdominal procedures.
MATERIALS AND METHODS A cross-sectional analysis of patients with a diagnosis of oral cavity, laryngeal, hypopharyngeal, or oropharyngeal cancer was performed using discharge data from the Nationwide Inpatient Sample (NIS), Healthcare Cost and Utilization Project, Agency for Healthcare Research and Quality. The NIS is the largest all-payer in-patient care database in the United States, containing data from approximately 8 million hospital stays each year from a stratified sample of 20% of non-federal U.S. hospitals from participating states.14 The NIS database provides information regarding the index hospital admission and includes patient demographic data, primary and secondary diagnoses, primary and secondary procedures, hospital HEAD & NECK—DOI 10.1002/HED
OCTOBER 2015
1509
GOOI ET AL.
characteristics, and in-patient and discharge mortality rates. The International Classification of Disease, 9th revision (ICD-9) codes were used to identify adult patients (18 years of age) who underwent an ablative procedure for a malignant oral cavity, laryngeal, hypopharyngeal, or oropharyngeal neoplasm for the years 1993 through 2010.15 Procedures that were unlikely to utilize reconstruction including excision/destruction of lesion, tonsillectomy, cordectomy, partial laryngectomy, and neck dissection as the index admission procedure when no other ablative procedure was recorded were excluded from analysis. Reconstructive procedures were obtained from codes for pedicled or free flap reconstruction (86.7, 86.70, 86.71, 86.72, 86.73, 86.74, 86.75, 86.8, and 86.89). Prior radiation was obtained from the codes for previous exposure to therapeutic or other ionizing radiation (V15.3). Comorbidity was graded using the Romano adaptation of the Charlson comorbidity index,16–18 excluding ICD-9 codes for the index cancer diagnosis from the solid tumor category. Cancer staging information is not available in the NIS, and, as a result, ICD-9 codes for metastases were excluded as these have not been shown to be a reliable surrogate for disease stage.19 Acute medical complications were derived from codes for acute cardiac events, acute pulmonary edema or failure, acute renal failure, acute hepatic failure, acute cerebrovascular events, sepsis, pneumonia, and urinary tract infection assigned at the time of hospital discharge, and surgical complications were derived from codes for complications directly resulting from surgical procedures assigned at the time of hospital discharge.15 Surgeon and hospital volume were modeled as categorical variables. The average annual number of head and neck cancer surgical cases performed per year of surgical activity was obtained by calculating the mean of the number of cases performed each year for each individual hospital or surgeon, for the years in which that hospital or surgeon performed at least one head and neck cancer surgery. Because, in 2003, the NIS instituted a change in the algorithm for creating a masked physician number that does not allow physicians to be tracked from before 2003 to after 2003, annual surgeon volumes were calculated separately for the time period before 2003 and from 2003 and beyond. Annual volumes were divided into quintiles, with high volume defined as those in the top volume quintile. Because volumes for the bottom 2 quintiles were similar, these were combined and, as a result, cutoff values for annual case volumes of 3, 4 to 7, 8 to 15, and >15 were used to classify surgeons by very low, low, intermediate, and high volume, and cutoff values of 10, 11 to 24, 25 to 49, and >50 were used to classify hospitals by very low, low, intermediate, and high volume. The primary clinical endpoints for analysis in this study were temporal trends in head and neck cancer reconstruction and variables associated with high-volume care. Flap reconstruction, in-hospital death, postoperative medical and surgical complications, length of hospitalization, and costs were examined as dependent variables. Secondary independent variables included in analyses were age, sex, race, payer source (commercial, health maintenance organization, Medicare, Medicaid, self-pay, or other), proce1510
HEAD & NECK—DOI 10.1002/HED
OCTOBER 2015
dure, nature of admission (emergent/urgent, or other), primary site, comorbidity, hospital ownership/control, hospital bed size, hospital location (rural or urban), geographic region, hospital teaching status, procedure, surgeon and hospital volume, and time interval. For statistical analysis of temporal trends, the study period was divided into 2 time intervals, 1993 to 2000 and 2001 to 2010, which was analyzed as an independent variable. American Joint Commission on Cancer tumor stage, tumor grade, histological subtype, and outcome after discharge were not available from the NIS database. Hospital-related charges for each index admission were examined for the time period 2001 to 2010, as hospitalrelated charges are not available from the NIS before 2001. Hospital-related charges were converted to the organizational cost of providing care using cost-to-charge ratios for individual hospitals. Cost-to-charge ratios were calculated using information from the detailed reports by hospitals to the Centers for Medicare and Medicaid Services, providing an estimate of the all-payer inpatient costto-charge ratio by hospital.20 This ratio was multiplied by each patient’s charge to obtain the cost per admission. All costs were adjusted for inflation based on U.S. Bureau of Labor Statistics indices, with results converted to 2012 U.S. dollars.21 To obtain national cost estimates, all discharges were re-weighted to account for cases in which cost estimates were missing.20 Data were analyzed using Stata 12 (StataCorp, College Station, TX). Associations between variables were analyzed using cross-tabulations, multivariate logistic regression, and multinomial logistic regression modeling. We conducted univariate and multivariate analysis to identify associations between dependent and independent variables. Non-zero discharge counts with fewer than 11 observations were masked in accordance with the Healthcare Cost and Utilization Project data use agreement. Data were weighted and modified hospital and discharge weights to correct for changes in sampling over time were applied. Variance estimation was performed using procedures for survey data analysis with replacement. Strata with one sampling unit were centered at the population mean. Variables with missing data for more than 10% of the population were coded with a dummy variable to represent the missing data in regression analysis. The primary clinical endpoints were evaluated using multiple and multivariate logistic regression analysis. Generalized linear regression modeling with a log link was used to analyze costs and length of stay because these variables were not normally distributed. This protocol was reviewed and approved by the Johns Hopkins Medical Institutions Institutional Review Board.
RESULTS There were 133,850 cases in 1993 to 2010, with 61,981 cases performed in 1993 to 2000 and 71,610 cases performed in 2001 to 2010 (Table 1). Flap reconstruction was performed in 6517 cases (11%) in 1993 to 2000 and in 11,505 cases (16%) in 2001 to 2010. The majority of patients were men, white, with commercial or Medicare insurance, and no or low comorbidity. The mean age was 62 years (range, 18–97 years), which did not differ by use of reconstruction. Mandibulectomy and total
HOSPITAL–ACQUIRED
CONDITIONS IN HEAD AND NECK SURGERY
TABLE 1. Demographic characteristics of patients from 1993 to 2000 and 2001 to 2010. No. of patients 1993–2000 N 5 61,981
Primary site Oral cavity Larynx Hypopharynx Oropharynx Age, y 40 40–64 65–80 >80 Race White Black Hispanic Asian or Pacific Islander Native American Other Unknown Sex Male Female Payor Private Medicare Medicaid Self-pay No charge Other Nature of admission Elective Emergency/urgent Comorbidity score 0 1 2 3 Procedure Partial glossectomy Total glossectomy Pharyngectomy Laryngectomy Maxillectomy Mandibulectomy Prior radiation Yes Hospital bed size Small Medium Large Hospital teaching status Non-teaching hospital Teaching hospital Hospital ownership Government, nonfederal Private, nonprofit Private, for profit Geographic region Northeast Midwest
No flap N 5 55,464
Flap N 5 6517
17.7% 56.0% 9.6% 16.7%
24.1% 19.4% 16.5% 40.0%
2.3% 51.8% 40.6% 5.3%
2.7% 55.8% 37.5% 4.0%
62.4% 10.5% 3.4% 0.7% 0.1% 1.1% 21.7%
68.5% 8.7% 4.4% 1.0% 0% 1.9% 15.5%
74.7% 25.3%
73.6% 26.4%
33.6% 45.6% 12.8% 4.3% 0.3% 3.3%
35.9% 42.6% 13.4% 3.6% 0.6% 3.9%
76.1% 23.9%
80.5% 19.5%
60.8% 28.7% 7.7% 2.8%
64.5% 16.0% 7.0% 2.5%
10.9% 4.1% 9.4% 67.7% 1.5% 21.5%
22.0% 14.1% 20.7% 43.9% 0.7% 36.9%
2.6%
4.5%
11.2% 17.6% 71.2%
14.2% 13.1% 72.7%
30.1% 69.9%
16.2% 83.8%
19.9% 74.0% 6.2%
29.5% 63.7% 6.8%
21.0% 23.9%
15.8% 21.2%
No. of patients 2001–2010 N 5 71,610 p value
No flap N 5 60,105
Flap N 5 11,505
Temporal trends in head and neck cancer surgery reconstruction Zhen Gooi, MBBS,1 Christine G. Gourin, MD, MPH,1,2* Kofi D. O. Boahene, MD,1 Patrick J. Byrne, MD,1 Jeremy D. Richmon, MD1 1
Department of Otolaryngology–Head and Neck Surgery, Johns Hopkins University, Baltimore, Maryland, 2Armstrong Institute for Patient Safety and Quality, Johns Hopkins University, Baltimore, Maryland.
Accepted 28 May 2014 Published online 21 July 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/hed.23786
ABSTRACT: Background. The purpose of this study was to analyze changing trends in head and neck cancer reconstructive surgery and analyze the effect of surgeon and hospital volume. Methods. Data from the Nationwide Inpatient Sample (NIS) for 133,850 patients who underwent a major ablative procedure for a malignant oral cavity, laryngeal, hypopharyngeal, or oropharyngeal neoplasm in 1993 to 2010 were analyzed using cross-tabulations and multivariate regression. Results. Reconstructive surgery in 2001 to 2010 was significantly associated with prior radiation (odds ratio [OR] 5 2.6; 95% confidence interval [95% CI] 5 1.4–4.9), comorbidity (OR 5 1.6; 95% CI 5 1.1–2.2), laryngeal cancer (OR 5 0.7; 95% CI 5 0.6–0.9), oropharyngeal cancer (OR 5 0.5; 95% CI 5 0.4–0.7), high-volume hospitals (OR 5 3.9; 95% CI 5 1.5–10.2), and high-volume surgeons (OR 5 2.0; 95% CI 5 1.1–
INTRODUCTION The oncologic resection of head and neck cancer incurs functional consequences to patients affecting speech, swallowing, and appearance. In efforts to restore normal physiology after surgical resection, a variety of techniques are used by reconstructive surgeons using pedicled or free tissue flaps. The safety and efficacy of these reconstructive techniques have been demonstrated in multiple large case series with low failure rates.1–5 Over the past 2 decades, there has been a paradigm shift in the primary treatment of laryngeal and oropharyngeal cancers, with increased utilization of chemoradiation compared to surgery.6–8 Consequently, the rate of salvage surgery for both laryngeal and oropharyngeal cancers nationally has been shown to be increasing, with a gradual shift in the performance of surgery to high volume centers and by high volume surgeons.9,10 It is currently unknown if the use of flap reconstruction in the era of increasing salvage surgery mirrors these patterns. Reconstruction of head and neck defects after oncologic resection, particularly those involving free tissue transfers, are complex procedures requiring advanced surgical training and close postoperative monitoring. The relationship between increased hospital volume, surgeon volume, and
*Corresponding author: C. G. Gourin, Johns Hopkins Outpatient Center, Department of Otolaryngology–Head and Neck Surgery, 601 N. Caroline Street, Suite 6260, Baltimore, MD 21287. E-mail: [email protected]
3.9), compared to 1993–2000. Reconstruction by high-volume surgeons was significantly associated with prior radiation (OR 5 1.8; 95% CI 5 1.1–3.1) and lower in-hospital mortality (OR 5 0.3; 95% CI 5 0.1– 1.0). A statistically significant negative correlation was observed between high-volume surgeons and length of hospitalization and hospital-related costs. Conclusion. These data reflect changing trends in head and neck cancer reconstructive surgery, with meaningful differences in the type of surgiC 2014 Wiley Periodicals, cal care provided by high-volume surgeons. V Inc. Head Neck 37: 1509–1517, 2015
KEY WORDS: hospital-acquired conditions, complications, surgery, head and neck neoplasms, costs, Nationwide Inpatient Sample
improved postoperative outcomes for complicated intrathoracic and intra-abdominal procedures measured through indices of short-term and long-term mortality have been demonstrated in a number of studies.11–13 We undertook this study to evaluate temporal trends in head and neck cancer reconstruction and variables associated with in-hospital mortality, postoperative complications, length of stay, and costs using a national hospital discharge database. We hypothesized that there is a trend toward centralization of care by higher volume surgeons and that this trend is associated with improved postoperative outcomes, similar to improved outcomes demonstrated for high-volume care for intrathoracic and intraabdominal procedures.
MATERIALS AND METHODS A cross-sectional analysis of patients with a diagnosis of oral cavity, laryngeal, hypopharyngeal, or oropharyngeal cancer was performed using discharge data from the Nationwide Inpatient Sample (NIS), Healthcare Cost and Utilization Project, Agency for Healthcare Research and Quality. The NIS is the largest all-payer in-patient care database in the United States, containing data from approximately 8 million hospital stays each year from a stratified sample of 20% of non-federal U.S. hospitals from participating states.14 The NIS database provides information regarding the index hospital admission and includes patient demographic data, primary and secondary diagnoses, primary and secondary procedures, hospital HEAD & NECK—DOI 10.1002/HED
OCTOBER 2015
1509
GOOI ET AL.
characteristics, and in-patient and discharge mortality rates. The International Classification of Disease, 9th revision (ICD-9) codes were used to identify adult patients (18 years of age) who underwent an ablative procedure for a malignant oral cavity, laryngeal, hypopharyngeal, or oropharyngeal neoplasm for the years 1993 through 2010.15 Procedures that were unlikely to utilize reconstruction including excision/destruction of lesion, tonsillectomy, cordectomy, partial laryngectomy, and neck dissection as the index admission procedure when no other ablative procedure was recorded were excluded from analysis. Reconstructive procedures were obtained from codes for pedicled or free flap reconstruction (86.7, 86.70, 86.71, 86.72, 86.73, 86.74, 86.75, 86.8, and 86.89). Prior radiation was obtained from the codes for previous exposure to therapeutic or other ionizing radiation (V15.3). Comorbidity was graded using the Romano adaptation of the Charlson comorbidity index,16–18 excluding ICD-9 codes for the index cancer diagnosis from the solid tumor category. Cancer staging information is not available in the NIS, and, as a result, ICD-9 codes for metastases were excluded as these have not been shown to be a reliable surrogate for disease stage.19 Acute medical complications were derived from codes for acute cardiac events, acute pulmonary edema or failure, acute renal failure, acute hepatic failure, acute cerebrovascular events, sepsis, pneumonia, and urinary tract infection assigned at the time of hospital discharge, and surgical complications were derived from codes for complications directly resulting from surgical procedures assigned at the time of hospital discharge.15 Surgeon and hospital volume were modeled as categorical variables. The average annual number of head and neck cancer surgical cases performed per year of surgical activity was obtained by calculating the mean of the number of cases performed each year for each individual hospital or surgeon, for the years in which that hospital or surgeon performed at least one head and neck cancer surgery. Because, in 2003, the NIS instituted a change in the algorithm for creating a masked physician number that does not allow physicians to be tracked from before 2003 to after 2003, annual surgeon volumes were calculated separately for the time period before 2003 and from 2003 and beyond. Annual volumes were divided into quintiles, with high volume defined as those in the top volume quintile. Because volumes for the bottom 2 quintiles were similar, these were combined and, as a result, cutoff values for annual case volumes of 3, 4 to 7, 8 to 15, and >15 were used to classify surgeons by very low, low, intermediate, and high volume, and cutoff values of 10, 11 to 24, 25 to 49, and >50 were used to classify hospitals by very low, low, intermediate, and high volume. The primary clinical endpoints for analysis in this study were temporal trends in head and neck cancer reconstruction and variables associated with high-volume care. Flap reconstruction, in-hospital death, postoperative medical and surgical complications, length of hospitalization, and costs were examined as dependent variables. Secondary independent variables included in analyses were age, sex, race, payer source (commercial, health maintenance organization, Medicare, Medicaid, self-pay, or other), proce1510
HEAD & NECK—DOI 10.1002/HED
OCTOBER 2015
dure, nature of admission (emergent/urgent, or other), primary site, comorbidity, hospital ownership/control, hospital bed size, hospital location (rural or urban), geographic region, hospital teaching status, procedure, surgeon and hospital volume, and time interval. For statistical analysis of temporal trends, the study period was divided into 2 time intervals, 1993 to 2000 and 2001 to 2010, which was analyzed as an independent variable. American Joint Commission on Cancer tumor stage, tumor grade, histological subtype, and outcome after discharge were not available from the NIS database. Hospital-related charges for each index admission were examined for the time period 2001 to 2010, as hospitalrelated charges are not available from the NIS before 2001. Hospital-related charges were converted to the organizational cost of providing care using cost-to-charge ratios for individual hospitals. Cost-to-charge ratios were calculated using information from the detailed reports by hospitals to the Centers for Medicare and Medicaid Services, providing an estimate of the all-payer inpatient costto-charge ratio by hospital.20 This ratio was multiplied by each patient’s charge to obtain the cost per admission. All costs were adjusted for inflation based on U.S. Bureau of Labor Statistics indices, with results converted to 2012 U.S. dollars.21 To obtain national cost estimates, all discharges were re-weighted to account for cases in which cost estimates were missing.20 Data were analyzed using Stata 12 (StataCorp, College Station, TX). Associations between variables were analyzed using cross-tabulations, multivariate logistic regression, and multinomial logistic regression modeling. We conducted univariate and multivariate analysis to identify associations between dependent and independent variables. Non-zero discharge counts with fewer than 11 observations were masked in accordance with the Healthcare Cost and Utilization Project data use agreement. Data were weighted and modified hospital and discharge weights to correct for changes in sampling over time were applied. Variance estimation was performed using procedures for survey data analysis with replacement. Strata with one sampling unit were centered at the population mean. Variables with missing data for more than 10% of the population were coded with a dummy variable to represent the missing data in regression analysis. The primary clinical endpoints were evaluated using multiple and multivariate logistic regression analysis. Generalized linear regression modeling with a log link was used to analyze costs and length of stay because these variables were not normally distributed. This protocol was reviewed and approved by the Johns Hopkins Medical Institutions Institutional Review Board.
RESULTS There were 133,850 cases in 1993 to 2010, with 61,981 cases performed in 1993 to 2000 and 71,610 cases performed in 2001 to 2010 (Table 1). Flap reconstruction was performed in 6517 cases (11%) in 1993 to 2000 and in 11,505 cases (16%) in 2001 to 2010. The majority of patients were men, white, with commercial or Medicare insurance, and no or low comorbidity. The mean age was 62 years (range, 18–97 years), which did not differ by use of reconstruction. Mandibulectomy and total
HOSPITAL–ACQUIRED
CONDITIONS IN HEAD AND NECK SURGERY
TABLE 1. Demographic characteristics of patients from 1993 to 2000 and 2001 to 2010. No. of patients 1993–2000 N 5 61,981
Primary site Oral cavity Larynx Hypopharynx Oropharynx Age, y 40 40–64 65–80 >80 Race White Black Hispanic Asian or Pacific Islander Native American Other Unknown Sex Male Female Payor Private Medicare Medicaid Self-pay No charge Other Nature of admission Elective Emergency/urgent Comorbidity score 0 1 2 3 Procedure Partial glossectomy Total glossectomy Pharyngectomy Laryngectomy Maxillectomy Mandibulectomy Prior radiation Yes Hospital bed size Small Medium Large Hospital teaching status Non-teaching hospital Teaching hospital Hospital ownership Government, nonfederal Private, nonprofit Private, for profit Geographic region Northeast Midwest
No flap N 5 55,464
Flap N 5 6517
17.7% 56.0% 9.6% 16.7%
24.1% 19.4% 16.5% 40.0%
2.3% 51.8% 40.6% 5.3%
2.7% 55.8% 37.5% 4.0%
62.4% 10.5% 3.4% 0.7% 0.1% 1.1% 21.7%
68.5% 8.7% 4.4% 1.0% 0% 1.9% 15.5%
74.7% 25.3%
73.6% 26.4%
33.6% 45.6% 12.8% 4.3% 0.3% 3.3%
35.9% 42.6% 13.4% 3.6% 0.6% 3.9%
76.1% 23.9%
80.5% 19.5%
60.8% 28.7% 7.7% 2.8%
64.5% 16.0% 7.0% 2.5%
10.9% 4.1% 9.4% 67.7% 1.5% 21.5%
22.0% 14.1% 20.7% 43.9% 0.7% 36.9%
2.6%
4.5%
11.2% 17.6% 71.2%
14.2% 13.1% 72.7%
30.1% 69.9%
16.2% 83.8%
19.9% 74.0% 6.2%
29.5% 63.7% 6.8%
21.0% 23.9%
15.8% 21.2%
No. of patients 2001–2010 N 5 71,610 p value
No flap N 5 60,105
Flap N 5 11,505
