Abstract
Background
In 2010, 158 000 cholecystectomies and 207 000 herniotomies (without bowel surgery) were performed in Germany as inpatient procedures, generally on a routine, elective basis. Deaths following such operations are rare events. We studied the potential association of death after cholecystectomy or herniotomy with risk factors that could have been detected beforehand, and we examined the types of complications that were documented in these cases.
Methods
Using nationwide hospital discharge data (DRG statistics) for the years 2009–2013, we analyzed the characteristics of patients who died in the hospital after undergoing a cholecystectomy for cholelithiasis or the repair of an inguinal, femoral, umbilical, or abdominal wall hernia. We compared these data with those of patients who survived and studied the impact of the coded comorbidities on the risk of death.
Results
In Germany, in the years 2009–2013, there were 2957 deaths after a total of 731 000 cholecystectomies (in-hospital mortality, 0.4%) and 1316 deaths after a total of 1 023 000 herniotomies without bowel surgery (0.13%). The patients who died were markedly older than those who did not, and they more commonly had comorbidities. Factors associated with a higher risk of death were age over 65 years, and comorbidities such as congestive heart failure, chronic pulmonary or hepatic disease, or poor nutritional status. Complications were coded much more often for the patients who died than for those who did not.
Conclusion
These findings suggest that there is potential for improvement in preoperative risk identification, complication avoidance, and the early recognition and treatment of complications, as well as in safe surgical technique. Measures to lower the mortality associated with herniotomy and cholecystectomy would lessen patients’ individual risk and thereby improve patient safety.
Gallbladder removal (cholecystectomy) and surgical repair of an inguinal, femoral, umbilical, or abdominal hernia (herniotomy) are common surgeries. A previous analysis found that approximately 158 000 inpatient cholecystectomies and 207 000 herniotomies without concomitant bowel surgery were performed in Germany in 2010. The median annual case number per hospital was 130 cholecystectomies and 156 herniotomies (1).
Death associated with this usually routine, scheduled surgery is perceived as a rare event. In Germany, the risk of dying in hospital in connection with a cholecystectomy was 0.49% (corresponding to one death per 204 operations) in 2010. For herniotomies the hospital mortality rate was 0.13% (corresponding to one death per 759 operations) (1). This means that, statistically, in an average hospital there is one cholecystectomy-related death every 19 months and one herniotomy-related death only once every five years. This makes it difficult to identify systematic correlations in everyday clinical practice.
Information obtained from peer reviews on quality improvement which analyzed deaths associated with cholecystectomy and herniotomy indicates that there is potential for improvement in indication and in perioperative and postoperative management (2, 3). In particular, patients at increased risk of complications as a result of comorbidities should be identified as high-risk patients before surgery and be provided with appropriate care (2).
The nationwide German hospital discharge data forms a complete dataset and can therefore be used to analyze even rare events. On the basis of this data, this study investigates the characteristics of patients who died in hospital in connection with a cholecystectomy or herniotomy and compares them to the characteristics of surviving patients.
It also analyzes the effect of documented comorbidities illnesses on the risk of death and examines complications and specific interventions during the course of treatment.
Methods
Data
Using controlled remote data processing, we analyzed the nationwide Diagnosis-Related Groups Statistics (DRG Statistics) provided by the German Federal Statistical Office’s Research Data Center (4). The nationwide DRG Statistics contain the data records of all acute inpatient hospital cases reimbursed according to the DRG system. This information includes age and sex, diagnoses (coded according to ICD-10-GM International Classification of Diseases, 10th Revision, German Modification) and procedures (coded according to the Surgery and Procedure Coding System [OPS, Operationen- und Prozedurenschlüssel]), hours of mechanical ventilation, and discharge mode for every inpatient case.
In order for analysis to cover sufficient deaths to provide statistical power, data for the years 2009 to 2013 were included and analyzed cumulatively.
Case definition
The units of analysis are inpatient hospital cases in which a patient underwent a cholecystectomy or herniotomy. Cholecystectomies include gallbladder removal for gallstones (cholelithiasis), excluding malignant neoplasias, extended cholecystectomies, and those performed concomitantly with other surgeries were excluded. Herniotomies are defined as surgeries for inguinal, femoral, umbilical, or abdominal hernia, excluding those with concomitant bowel surgery (eTable 1). Analyses are limited to patients aged 20 years or over.
eTable 1. Case definition*.
| Criterion | Principal diagnosis (ICD-10) | Secondary diagnosis (ICD-10) | Procedure (OPS) | |
|---|---|---|---|---|
| Cholecystectomy for cholelithiasis; excluding extended cholecystectomies or cholecystectomies performed concomitantly with other surgeries | Inclusion | K80 | 5–511.0, 5–511.1, 5–511.2, 5–511.x, 5–511.y | |
| Exclusion | C00–C97 | 5–511.3, 5–511.4, 5–511.5 | ||
| Herniotomy for inguinal, femoral, umbilical, or abdominal hernia without concomitant bowel surgery | Inclusion | K40, K41, K42, K43 | 5–530, 5–531, 5–534, 5–535, 5–536 | |
| Exclusion | 5–451, 5–452, 5–453, 5–454, 5–455, 5–456, 5–458, 5–459, 5–460, 5–461, 5–462, 5–463, 5–464, 5–465, 5–466, 5–467, 5–468, 5–469, 5–484, 5–485, 5–530.4, 5–530.8, 5–531.4, 5–531.8 |
*The analyses performed for this study are limited to patients aged 20 years or over.
OPS: Classification of Operations and Procedures (the German modification of the International Classification of Procedures in Medicine)
Comorbidities, complications, specific interventions
Comorbidities are taken to include conditions that probably existed prior to inpatient admission (e.g. heart failure, chronic lung disease) and that represent potential risk factors for in-hospital death (5– 9). Complications include events that are classified as complications in the ICD (e.g. surgical complications such as accidental perforation and suture failure) and serious conditions that would contraindicate the surgery in question (e.g. pulmonary embolism, acute kidney failure) and therefore probably first occurred during treatment. Comorbidities and complications were identified using the coded diagnoses (eTable 2).
eTable 2. Definition of variables.
| Comorbidity | ICD-10 code | Comments |
| Cardiac arrhythmia | I44.2, I48, Z45.0, Z95.0 | |
| Heart failure/cardiomyopathy | I50, I11.0, I13.0, I13.2, I42.0, I42.6, I42.7, I42.8, I42.9 | |
| Chronic ischemic heart disease | I25 | |
| Hypertension (without heart or kidney failure) | I10, I11.9, I12.9, I13.9, I15 | |
| Aortic/mitral valve defects | I34.0, I34.2, I35.0, I35.1, I35.2, I05.0, I05.1, I05.2, I06.0, I06.1, I06.2, Q23.1, Q23.2, Q23.3 | |
| Atherosclerosis of the extremities | I70.2 | |
| Chronic lung disease | J41, J42, J44, J45, J47 | |
| Chronic liver disease | B18, I86.4, I98.2, K70, K73, K74, K76.0, K76.1, K76.5, K76.6, K76.7 | |
| Chronic pancreatitis | K86.0, K86.1 | |
| Severe kidney disease | N03, N04, N05, N07, N08, N11, N12, N14, N15, N16 | |
| Chronic kidney failure | I12.0, I13.1, I13.2, N18, N19, Z99.2 | |
| Diabetes mellitus | E10, E11, E12, E13, E14 | |
| Obesity | E66 | |
| Cachexia/malnutrition | R64, R63.4, E43, E44 | |
| Peritoneal adhesions | K66.0 | Cholecystectomy only |
| Coagulation disorder | D66, D67, D68, D69.1, D69.3, D69.4 | |
| Acute cholecystitis, cholangitis, or biliary pancreatitis | K80.0, K80.3, K81.0, K83.0, K85.1 | Cholecystectomy only |
| Malignant neoplasia | C00-C97 | Herniotomy only |
| Complication | ICD-10 code | Comments |
| Deep vein thrombosis/pulmonary embolism | I80.1, I80.2, I80.3, I82.2, I26 | |
| Myocardial infarction/TIA/stroke | I21, I22, G45, I60, I61, I63, I64 | |
| Cardiogenic shock/cardiac arrest/ventricular fibrillation | R57.0, I46, I49.0 | |
| Other shock (e.g. hypovolemic, anaphylactic) | R57.1, R57.8, R57.9, T81.1, T88.2, T88.6 | |
| Acute kidney failure | N17, N99.0 | |
| Pneumonia | A48.1, J10.0, J11.0, J12, J13, J14, J15, J16, J17, J18, J69 | |
| Wound infection | T81.4 | |
| Sepsis/SIRS | A40, A41, R57.2, R65 | |
| Postoperative intestinal obstruction | K91.3 | |
| Perioperative or postoperative surgical complication | T81.7, T81.8, T81.9, T88.7, Y57.9 | |
| Anesthesia-related complication | T88.3, T88.4, T88.5, T41 | |
| Other complications associated with medical procedures | I97.8, I97.9, T80, T81.8, T81.9, T88.7 | |
| Coagulation complication | D65.1, D69.53 | |
| Peritonitis | K65 | Herniotomy only |
| Intervention | Classification of Operations and Procedures (OPS) code | Comments |
| Intensive care | 8–980 | |
| Blood transfusion (whole blood or red cell concentrate, six units or more) | 8–800.1, 8–800.c1 bis 8–800.cr, 8–800.70 bis 8–800.7e, 8–800.7g, 8–800.h | Change in OPS 2010 |
| Resuscitation | 8–771, 8–772, 8–773 | |
| Relaparotomy, repeat surgery | 5–541.2, 5–983 | |
| Switch from laparoscopic to open surgery | 5–511.21, 5–511.22 | Cholecystectomy only |
| Clinical or quality-assurance autopsy | 9–990 | |
In addition, specific interventions that indicate complicated patient progress were also investigated. These include, for example, blood transfusions of more than six units and relaparatomies, identified using coded procedures (eTable 2). Inpatients receiving mechanical ventilation for longer than 24 hours were also identified, on the basis of documented mechanical ventilation duration.
Analysis
Patient characteristics, comorbidities, complications, and interventions were stratified by survival status. The effects of age, sex, and comorbidities on the in-hospital mortality risk were analyzed using odds ratios that estimate the risk of death when a characteristic (e.g. a comorbidity) is present in relation to the risk of death when the characteristic is not present. In addition to crude odds ratios, odds ratios adjusted for age, sex, and comorbidities are also provided. These were calculated using generalized logistic regression (10). The adjusted odds ratios quantify the change in risk independently of the effects of other factors in the regression equation. The statistical significance of the odds ratios was assessed using 95% confidence intervals. The discriminatory power of the regression models was assessed using the c-statistic, which is equivalent to the area under the ROC (receiver operating characteristic) curve (11, 12). This measure describes the association of deaths predicted by the regression model and observed deaths. c-values can range between 0.5 (no discrimination) and 1.0 (perfect discrimination). All calculations were performed using SAS version 9.3.
Results
Patient characteristics
Approximately 731 000 cases of cholecystectomy for cholelithiasis and 1 023 000 cases of herniotomy without concomitant bowel surgery were identified during the study period. Between 2009 and 2013, the in-hospital mortality rate for cholecystectomy remained steady at 0.4%. For herniotomy in-hospital mortality was 0.14% from 2009 to 2011, 0.11% in 2012, and 0.13% in 2013. A total of 2957 deaths occurred following cholecystectomy and 1316 following herniotomy during the study period (Table 1).
Table 1. Patient characteristics by year.
| 2009 | 2010 | 2011 | 2012 | 2013 | Total | |
|---|---|---|---|---|---|---|
| Cholecystectomy for cholelithiasis | ||||||
| No. of cases | 145670 | 146041 | 146178 | 146726 | 146811 | 731426 |
| No. of deaths | 626 | 576 | 595 | 578 | 582 | 2957 |
| In-hospital mortality rate | 0.43% | 0.39% | 0.41% | 0.39% | 0.40% | 0.40% |
| Mean age (years) | 55.8 | 55.9 | 55.9 | 56.0 | 55.9 | 55.9 |
| Percentage aged 65 years or over | 35.3% | 34.8% | 34.1% | 34.1% | 33.6% | 34.4% |
| Percentage female | 66.9% | 66.3% | 65.9% | 65.4% | 65.1% | 65.9% |
| Mean no. of secondary diagnoses coded per case | 2.8 | 2.9 | 3.1 | 3.2 | 3.3 | 3.1 |
| Mean no. of investigated comorbidities per case | 1.3 | 1.3 | 1.3 | 1.4 | 1.4 | 1.3 |
| No. of hospitals | 1139 | 1125 | 1103 | 1080 | 1062 | 1206 |
| Mean no. of cases per hospital | 128 | 130 | 133 | 136 | 138 | 133 |
| Herniotomy for inguinal, femoral, umbilical, or abdominal hernia | ||||||
| No. of cases | 201796 | 200969 | 205699 | 208930 | 205488 | 1022882 |
| No. of deaths | 281 | 273 | 281 | 220 | 261 | 1316 |
| In-hospital mortality rate | 0.14% | 0.14% | 0.14% | 0.11% | 0.13% | 0.13% |
| Mean age (years) | 59.7 | 59.6 | 59.7 | 59.7 | 59.6 | 59.7 |
| Percentage aged 65 years or over | 44.4% | 43.5% | 43.0% | 42.5% | 42.0% | 43.1% |
| Percentage female | 20.8% | 20.5% | 20.3% | 20.2% | 20.0% | 20.3% |
| Mean no. of secondary diagnoses coded per case | 2.2 | 2.2 | 2.3 | 2.4 | 2.5 | 2.3 |
| Mean no. of investigated comorbidities per case | 0.8 | 0.8 | 0.8 | 0.9 | 0.9 | 0.8 |
| No. of hospitals | 1202 | 1195 | 1158 | 1137 | 1116 | 1294 |
| Mean no. of cases per hospital | 168 | 168 | 178 | 184 | 184 | 176 |
Stratified analysis of the characteristics of surviving and deceased patients shows, unsurprisingly, that for both procedures those who died were substantially older than those who survived. For cholecystectomy the proportion of females was slightly lower among those who died than among those who survived; for herniotomies it was somewhat higher. Almost all comorbidities analyzed were recorded more frequently among those who died than among those who survived. At least one of the analyzed comorbidities had been coded in 99% of deaths following cholecystectomy and 91% of deaths following herniotomy. The mean duration of hospital stay was also substantially longer for patients who died (Table 2).
Table 2. Patient characteristics by survival status.
| Cholecystectomy for cholelithiasis | Herniotomy for inguinal, femoral, umbilical, or abdominal hernia | |||||||
|---|---|---|---|---|---|---|---|---|
| Surviving | Dying | Surviving | Dying | |||||
| Cases | Percentage | Cases | Percentage | Cases | Percentage | Cases | Percentage | |
| Total | 728469 | 100.0% | 2957 | 100.0% | 1021566 | 100.0% | 1316 | 100.0% |
| Aged 65 years or over | 248841 | 34.2% | 2759 | 93.3% | 439342 | 43.0% | 1160 | 88.1% |
| Female | 480496 | 66.0% | 1540 | 52.1% | 207576 | 20.3% | 515 | 39.1% |
| Investigated comorbidities | ||||||||
| Cardiac arrhythmia | 37261 | 5.1% | 1249 | 42.2% | 64084 | 6.3% | 468 | 35.6% |
| Heart failure/dilated cardiomyopathy | 23478 | 3.2% | 1282 | 43.4% | 34331 | 3.4% | 500 | 38.0% |
| Chronic ischemic heart disease | 39402 | 5.4% | 791 | 26.8% | 80765 | 7.9% | 306 | 23.3% |
| Hypertension (without heart or kidney failure) | 251753 | 34.6% | 1453 | 49.1% | 339720 | 33.3% | 574 | 43.6% |
| Aortic/mitral valve defects | 4246 | 0.6% | 130 | 4.4% | 8160 | 0.8% | 58 | 4.4% |
| Atherosclerosis of the extremities | 4293 | 0.6% | 159 | 5.4% | 8874 | 0.9% | 63 | 4.8% |
| Chronic lung disease | 32971 | 4.5% | 426 | 14.4% | 57997 | 5.7% | 213 | 16.2% |
| Chronic liver disease | 22941 | 3.1% | 321 | 10.9% | 11319 | 1.1% | 164 | 12.5% |
| Chronic pancreatitis | 1639 | 0.2% | 21 | 0.7% | 991 | 0.1% | 5 | 0.4% |
| Severe kidney disease | 4723 | 0.6% | 259 | 8.8% | 3849 | 0.4% | 62 | 4.7% |
| Chronic kidney failure | 26449 | 3.6% | 1068 | 36.1% | 31078 | 3.0% | 394 | 29.9% |
| Diabetes mellitus | 73727 | 10.1% | 1050 | 35.5% | 88416 | 8.7% | 307 | 23.3% |
| Obesity | 90628 | 12.4% | 259 | 8.8% | 76266 | 7.5% | 118 | 9.0% |
| Cachexia/malnutrition | 1320 | 0.2% | 72 | 2.4% | 1337 | 0.1% | 51 | 3.9% |
| Coagulation disorder | 15929 | 2.2% | 891 | 30.1% | 13995 | 1.4% | 265 | 20.1% |
| Peritoneal adhesions*1 | 85749 | 11.8% | 604 | 20.4% | – | – | – | – |
| Acute cholecystitis, cholangitis, or biliary pancreatitis | 240599 | 33.0% | 2295 | 77.6% | – | – | – | – |
| Malignant neoplasia*2 | – | – | – | – | 12565 | 1.2% | 122 | 9.3% |
| At least one investigated comorbidity | 485814 | 66.7% | 2918 | 98.7% | 473034 | 46.3% | 1200 | 91.2% |
| Mean length of hospital stay (days) | 5.9 | 17.4 | 3.1 | 11.9 | ||||
*1As the case definition for herniotomy excludes cases with adhesiolysis (because it excludes concomitant bowel surgery), peritoneal adhesions are not included as a secondary diagnosis.
*2The case definition for cholecystectomy excludes cases with malignant neoplasia as a secondary diagnosis.
The corresponding crude mortality rates are shown in eTable 3
Factors associated with the risk of death following cholecystectomy
For cholecystectomy, the crude figures indicate that ages over 65 years are associated with a risk of death 27 times higher than that of younger patients. After adjustment for sex and comorbidities, the risk of death for patients aged over 65 remains 10 times as high as for younger patients, independently of other factors. According to the crude figures, female sex appears to be associated with a lower risk of death, but this correlation is no longer discernible after adjustment (Table 3).
Table 3. Crude and adjusted odds ratios for in-hospital death.
| Cholecystectomy for cholelithiasis | Herniotomy for inguinal, femoral, umbilical, or abdominal hernia | |||||||
|---|---|---|---|---|---|---|---|---|
| Crude OR | 95% CI | Adjusted OR | 95% CI | Crude OR | 95% CI | Adjusted OR | 95% CI | |
| Aged 65 years or over | 26.9 | (23.2 to 31.0) | 10.0 | (8.5 to 11.7) | 9.9 | (8.3 to 11.6) | 5.2 | (4.4 to 6.2) |
| Female | 0.6 | (0.5 to 0.6) | 1.1 | (1.0 to 1.2) | 2.5 | (2.3 to 2.8) | 2.3 | (2.0 to 2.6) |
| Comorbidities | ||||||||
| Cardiac arrhythmia | 13.6 | (12.6 to 14.6) | 1.7 | (1.6 to 1.9) | 8.2 | (7.4 to 9.2) | 1.7 | (1.4 to 2.0) |
| Heart failure/dilated cardiomyopathy | 23.0 | (21.3 to 24.7) | 3.6 | (3.3 to 4.0) | 17.6 | (15.8 to 19.7) | 4.3 | (3.6 to 5.0) |
| Chronic ischemic heart disease | 6.4 | (5.9 to 6.9) | 1.2 | (1.1 to 1.3) | 3.5 | (3.1 to 4.0) | 1.1 | (0.9 to 1.3) |
| Hypertension (without heart or kidney failure) | 1.8 | (1.7 to 2.0) | 0.7 | (0.6 to 0.8) | 1.6 | (1.4 to 1.7) | 0.7 | (0.6 to 0.8) |
| Aortic/mitral valve defects | 7.8 | (6.6 to 9.4) | 1.2 | (1.0 to 1.5) | 5.7 | (4.4 to 7.5) | 1.1 | (0.8 to 1.5) |
| Atherosclerosis of the extremities | 9.6 | (8.1 to 11.3) | 1.7 | (1.4 to 2.0) | 5.7 | (4.5 to 7.4) | 1.8 | (1.3 to 2.4) |
| Chronic lung disease | 3.6 | (3.2 to 3.9) | 1.3 | (1.2 to 1.5) | 3.2 | (2.8 to 3.7) | 1.5 | (1.3 to 1.8) |
| Chronic liver disease | 3.7 | (3.3 to 4.2) | 2.1 | (1.8 to 2.4) | 12.7 | (10.8 to 15) | 5.9 | (4.9 to 7.1) |
| Chronic pancreatitis | 3.2 | (2.1 to 4.9) | 2.1 | (1.3 to 3.3) | 3.9 | (1.6 to 9.5) | 1.9 | (0.8 to 4.7) |
| Severe kidney disease | 14.7 | (12.9 to 16.8) | 1.2 | (1.0 to 1.4) | 13.1 | (10.1 to 16.9) | 1.0 | (0.7 to 1.5) |
| Chronic kidney failure | 15.0 | (13.9 to 16.2) | 2.2 | (2.0 to 2.5) | 13.6 | (12.1 to 15.3) | 2.7 | (2.3 to 3.2) |
| Diabetes mellitus | 4.9 | (4.5 to 5.3) | 1.4 | (1.3 to 1.6) | 3.2 | (2.8 to 3.6) | 1.3 | (1.1 to 1.5) |
| Obesity | 0.7 | (0.6 to 0.8) | 0.6 | (0.5 to 0.7) | 1.2 | (1.0 to 1.5) | 1.0 | (0.8 to 1.2) |
| Cachexia/malnutrition | 13.7 | (10.8 to 17.5) | 3.5 | (2.6 to 4.7) | 30.8 | (23.1 to 40.9) | 8.0 | (5.5 to 11.7) |
| Coagulation disorder | 19.3 | (17.8 to 20.9) | 3.5 | (3.1 to 3.9) | 18.2 | (15.8 to 20.8) | 4.7 | (3.9 to 5.7) |
| Peritoneal adhesions*1 | 1.9 | (1.8 to 2.1) | 1.3 | 1.2 to 1.5) | to | to | to | to |
| Acute cholecystitis, cholangitis, or biliary pancreatitis | 7.0 | (6.4 to 7.7) | 3.2 | (2.9 to 3.5) | to | to | to | to |
| Malignant neoplasia*2 | – | – | – | – | 8.2 | (6.8 to 9.9) | 4.1 | (3.3 to 5.0) |
| Association between predicted probability of death and observed mortality | c=0.925 | c=0.847 | ||||||
OR: odds ratio; CI: confidence interval
*1As the case definition for herniotomy excludes cases with adhesiolysis (because it excludes concomitant bowel surgery), peritoneal adhesions are not included as a secondary diagnosis.
*2The case definition for cholecystectomy excludes cases with malignant neoplasia as a secondary diagnosis.
The crude odds ratios of all analyzed comorbidities other than obesity indicate an association with an increased risk of death. After adjustment, an increased risk independent of other factors remains for most comorbidities. The strongest association occurs in patients with heart failure or dilated cardiomyopathy, whose risk of death is 3.6 times higher than that of patients for whom no such comorbidity is coded. Patients with malnutrition and those with coagulation disorders both have a 3.5-fold risk increase. Inflammatory conditions at the surgical site (cholangitis, acute cholecystitis, or biliary pancreatitis) are associated with 3.2-fold increase in the risk of death. In patients with chronic liver disease, chronic pancreatitis, or chronic kidney failure the risk of death is more than twice as high as in patients without these comorbidities. Other comorbidities associated with a significantly increased risk of death according to the adjusted odds ratios are cardiac arrhythmias, chronic ischemic heart disease, atherosclerosis, chronic lung disease, diabetes mellitus, and peritoneal adhesions. For obesity, both the crude and the adjusted analysis actually indicate a protective effect. This means that cholecystectomy patients for whom obesity is coded as a secondary diagnosis have a lower risk of dying in the hospital. Following adjustment, patients with hypertension also have a lower risk of death than patients for whom it is not coded. Aortic or mitral valve defects and severe kidney disease showed no significant effect on the risk of death after adjustment (Table 3).
The c-value of the regression model was 0.925. This means that the model has a high power of discrimination and can differentiate very well between patients who died and those who survived on the basis of the variables used here.
Factors associated with the risk of death following herniotomy
In herniotomy patients, after adjustment age over 65 years is associated with a more than 5-fold increase in the risk of death. Women have more than twice the risk of dying in the hospital of men according to both crude and adjusted figures (table 3).
According to the crude figures, all comorbidities appear to be associated with an increased risk of death. After adjustment a greatly increased risk remains for several comorbidities: the risk of death in malnourished patients is eight times higher than in those for whom malnutrition is not coded. Chronic liver disease is associated with an almost six-fold increase in risk. There is a more than four-fold increase in risk associated with coagulation disorders, malignant neoplasias, and heart failure. For patients with chronic kidney failure the risk of death is 2.7 times higher. Other significant increases in risk are found in patients with cardiac arrhythmias, atherosclerosis, chronic lung disease, and diabetes mellitus.
After adjustment, the increase in risk for chronic pancreatitis is no longer significant. No effect on the risk of death was identified for chronic ischemic heart disease, aortic or mitral valve defects, severe kidney disease, or obesity. The secondary diagnosis hypertension is associated with a reduced risk of death according to adjusted figures (Table 3).
The regression model has a c-value of 0.847, indicating a very good power of discrimination.
Complications and specific interventions
All the complications investigated were coded more frequently for patients who died than for those who survived. Sepsis was recorded for 37% of deceased cholecystectomy patients and 17% of deceased herniotomy patients. Other common complications were acute kidney failure (37% and 29% respectively), pneumonia (29% and 27%), cardiogenic shock (17% and 21%), other shock (11% and 9%), and cardiovascular or cerebrovascular events such as myocardial infarction or stroke (9% and 11%). Surgical complications (e.g. hemorrhage, accidental perforation, suture failure) were recorded in 19% of deaths following cholecystectomy and 13% of deaths following herniotomy. In total, at least one of the investigated complications was coded for 83% of deaths following cholecystectomy and 78% of deaths following herniotomy (Table 4).
Table 4. Complications and specific interventions by survival status.
| Cholecystectomy for cholelithiasis | Herniotomy for inguinal, femoral, umbilical, or abdominal hernia | |||||||
|---|---|---|---|---|---|---|---|---|
| Surviving | Dying | Surviving | Dying | |||||
| Cases | Percentage | Cases | Percentage | Cases | Percentage | Cases | Percentage | |
| Total | 728469 | 100% | 2957 | 100% | 1021566 | 100% | 1316 | 100% |
| Complications | ||||||||
| Deep vein thrombosis/pulmonary embolism | 704 | 0.1% | 106 | 3.6% | 697 | 0.1% | 101 | 7.7% |
| Myocardial infarction/TIA/stroke | 1577 | 0.2% | 274 | 9.3% | 1260 | 0.1 % | 141 | 10.7% |
| Cardiogenic shock/cardiac arrest/ ventricular fibrillation | 534 | 0.1% | 512 | 17.3% | 467 | 0.05% | 272 | 20.7% |
| Other shock (e.g. hypovolemic, anaphylactic) | 1126 | 0.2% | 328 | 11.1% | 1335 | 0.1% | 122 | 9.3% |
| Acute kidney failure | 3027 | 0.4% | 1088 | 36.8% | 1407 | 0.1% | 377 | 28.6% |
| Pneumonia | 5045 | 0.7% | 872 | 29.5% | 2199 | 0.2% | 359 | 27.3% |
| Wound infection | 6905 | 0.9% | 193 | 6.5% | 4132 | 0.4% | 56 | 4.3% |
| Sepsis/SIRS | 3981 | 0.5% | 1095 | 37.0% | 414 | 0.04% | 222 | 16.9% |
| Perioperative or postoperative surgical complication | 18319 | 2.5% | 549 | 18.6% | 28292 | 2.8% | 168 | 12.8% |
| Anesthesia-related complication | 5249 | 0.7% | 58 | 2.0% | 6394 | 0.6% | 17 | 1.3% |
| Other complication associated with medical procedures | 16572 | 2.3% | 288 | 9.7% | 18005 | 1.8% | 133 | 10.1% |
| Coagulation complication (DIC or HIT II) | 371 | 0.1% | 120 | 4.1% | 215 | 0.02% | 27 | 2.1% |
| Peritonitis* | – | – | – | 706 | 0.1% | 82 | 6.2% | |
| Postoperative intestinal obstruction | 664 | 0.1% | 44 | 1.5% | 778 | 0.1% | 22 | 1.7% |
| At least one investigated complication | 52826 | 7.3% | 2445 | 82.7% | 60123 | 5.9% | 1030 | 78.3% |
| Specific interventions | ||||||||
| Ventilation for more than 24 hours | 2749 | 0.4% | 1292 | 43.7% | 806 | 0.1% | 342 | 26.0% |
| Intensive care | 12783 | 1.8% | 1491 | 50.4% | 4681 | 0.5% | 481 | 36.6% |
| Blood transfusion (at least 6 units) | 1481 | 0.2% | 519 | 17.6% | 553 | 0.1% | 132 | 10.0% |
| Resuscitation | 440 | 0.1% | 542 | 18.3% | 349 | 0.03% | 300 | 22.8% |
| Relaparotomy, repeat surgery | 3749 | 0.5% | 334 | 11.3% | 8543 | 0.8% | 60 | 4.6% |
| Switch from laparoscopic to open surgery | 29655 | 4.1% | 521 | 17.6% | – | – | – | – |
| Autopsy | – | – | 13 | 0.4% | – | – | 7 | 0.5% |
*Peritonitis is included as a complication in herniotomy only (for cholecystectomy it is unclear whether it was present before surgery or acquired during treatment).TIA: transient ischemic attack; SIRS: systemic inflammatory response syndrome; DIC: disseminated intravascular coagulation; HIT II: heparin-induced thrombocytopenia, type II
The Figure displays the crude mortality rates for complications. This shows that almost all the analyzed complications are associated with a substantially worse prognosis for survival. Mortality rates of more than 10% were found for patients with thrombotic, cardiovascular, cerebrovascular, or serious cardiac events; acute kidney failure; pneumonia; sepsis; coagulation-related complications; and peritonitis.
Figure.
Crude mortality rate stratified by complication. TIA: transient ischemic attack; SIRS: systemic inflammatory response syndrome; DIC: disseminated intravascular coagulation; HIT II: heparin-induced thrombocytopenia, type II *Peritonitis is included as a complication in herniotomy only (for cholecystectomy it is unclear whether it was present before surgery or acquired during treatment)
The analyzed interventions were also recorded more frequently in patients who died than in those who survived: intensive care was recorded for more than half of cholecystectomy patients who died and 37% of herniotomy patients who died. Ventilation lasting more than 24 hours was administered to 44% of those who died following cholecystectomy and 26% of those who died following herniotomy. Blood transfusions of more than six units, which provide indirect evidence of hemorrhage as a complication, were recorded for 18% of patients who died after cholecystectomy and 10% of patients who died after herniotomy. Relaparotomy or repeat surgery was coded for 11% of deaths following cholecystectomy and 5% of deaths following herniotomy. Resuscitation was performed on 18% of cholecystectomy patients who died and 23% of herniotomy patients who died.
Autopsies were documented for 0.4% of deaths following cholecystectomy and 0.5% of deaths following herniotomy (Table 4).
Discussion
Despite the rarity of these events, the nationwide German database used for this study allowed more than 4000 deaths to be analyzed and compared to 1.75 million surviving patients. This large number of cases made it possible to show common characteristics of patients who died and to investigate the effect of coded comorbidities on the risk of in-hospital death.
In this study comorbidities are identified solely on the basis of corresponding diagnoses being recorded in inpatient data records. As coding of a comorbidity implies that personnel were aware of it during clinical treatment, there may be potential for optimization using a risk-adapted approach in indication, preparation for surgery, and perioperative and postoperative care.
Most comorbidities examined here were associated with a significant increase in the risk of death. The effects of, for example, cardiovascular diseases; chronic lung, liver, or kidney disease; and malnutrition on the risk of complications or death following surgery, including visceral surgery, are well documented in the literature (7– 9). Identifying such risks is therefore a central part of preoperative preparation (13). Accurate medical history taking and careful physical examination are the basis for identification of diseases that have not previously been detected or have not been adequately treated and affect perioperative risk (14, 15). To the extent that these are factors that can be altered, and if surgery is not so urgent as to make this impossible, preoperative optimization can reduce the risk of complications. This concerns, for example, nutrition therapy for malnourished patients (16, 17) and optimization of treatment for heart failure (18).
Analysis of complications and interventions indicates that there may be potential for optimizing the management of complications (19). This concerns prevention or early identification and adequate treatment of nosocomial infections (20), cardiac events (21), kidney injury (22), and other conditions. In addition, the results of our study suggest that there are further aspects of patient safety in which action is required in German healthcare (23). For example, the frequency of coded surgical complications and blood transfusions in patients who died indicates that it may be possible to improve safety in surgery still further. An effective tool for preventing intraoperative complications and reducing mortality in patients undergoing surgery is surgical safety checklists (24), comprehensive use of which is recommended for all surgeries (25). Regular morbidity and mortality conferences at which the treatment team analyzes unexpected deaths or complications are also seen as the standard for perioperative medicine (26).
As far as can be determined on the basis of comparable publications, in-hospital mortality following cholecystectomy and herniotomy is no higher in Germany than in other industrialized countries. In-hospital mortality for laparoscopic cholecystectomy is 0.5% in the USA, for example (27). In Denmark, 30-day mortality rates of 0.35% for cholecystectomy (28) and 0.5% in herniotomy (29) were reported. The results of this analysis do indicate, however, that mortality related to these surgeries in Germany could be reduced. This is also suggested by the findings of statutory external quality assurance for cholecystectomy: in 2013 deaths were identified despite a low mortality risk in 169 hospitals (30).
Limitations
The characteristics analyzed here may be biased by erroneous or selective diagnosis coding. It is therefore possible, for example, that comorbidities are more likely to be perceived when treatment is complicated and are therefore coded more frequently in patients who died than in those who survived. This kind of bias would lead to an overestimate of the association between comorbidities and risk of death.
Conclusion
Deaths are strongly associated with coded comorbidities that are usually identifiable before surgery, even for routine surgeries such as cholecystectomy and herniotomy that are low in complexity and usually scheduled. Unlike in high-risk surgeries, because such deaths are rare in individual hospitals awareness of such patient-specific risk factors in such apparently simple surgeries may not always be sufficient. In addition to determining patient-specific risks on the basis of patients’ medical history as a requirement for risk-adapted treatment planning, preparation, and execution, there may be potential for improving the management of complications.
Reducing mortality for herniotomy and cholecystectomy would save only a few lives in absolute terms. For the patients, however, who usually consider a herniotomy or cholecystectomy low risk surgery, appropriate steps would contribute substantially to reducing their individual risk. In addition, measures to prevent deaths might reduce complication rates as well as mortality and therefore improve overall patient safety.
Key Messages.
Between 2009 in 2013 in Germany, there were 2957 deaths following cholecystectomy for gallstones (mortality: 0.4%) and 1316 deaths following herniotomy without concomitant bowel surgery (mortality: 0.13%).
In addition to higher age, coded comorbidities such as cardiovascular disease; chronic lung, liver, or kidney disease; and compromised nutritional status are associated with an increased risk of death.
Unsurprisingly, complications are coded substantially more frequently for patients who died than for survivors.
There may be potential for optimization in preoperative risk identification, prevention or early identification and treatment of complications, and safe surgery technique.
Reducing mortality for herniotomy and cholecystectomy would save only a few lives in absolute terms. However, appropriate steps would contribute substantially to reducing individual patient risk and thereby improve patient safety.
eTable 3. Crude mortality rates stratified by patient characteristics.
| Cholecystectomy for cholelithiasis | Herniotomy for inguinal, femoral, umbilical, or abdominal hernia | |||||
|---|---|---|---|---|---|---|
| Total cases | Deaths | Crude mortality rate | Total cases | Deaths | Crude mortality rate | |
| Total | 731426 | 2957 | 0.40% | 1022882 | 1316 | 0.13% |
| Aged 65 years or over | 251600 | 2759 | 1.10% | 440502 | 1160 | 0.26% |
| Female | 482036 | 1540 | 0.32% | 208091 | 515 | 0.25% |
| Investigated comorbidities | ||||||
| Cardiac arrhythmia | 38510 | 1249 | 3.24% | 64552 | 468 | 0.72% |
| Heart failure/dilated cardiomyopathy | 24760 | 1282 | 5.18% | 34831 | 500 | 1.44% |
| Chronic ischemic heart disease | 40193 | 791 | 1.97% | 81071 | 306 | 0.38% |
| Hypertension (without heart or kidney failure) | 253206 | 1453 | 0.57% | 340294 | 574 | 0.17% |
| Aortic/mitral valve defects | 4376 | 130 | 2.97% | 8218 | 58 | 0.71% |
| Atherosclerosis of the extremities | 4452 | 159 | 3.57% | 8937 | 63 | 0.70% |
| Chronic lung disease | 33397 | 426 | 1.28% | 58210 | 213 | 0.37% |
| Chronic liver disease | 23262 | 321 | 1.38% | 11483 | 164 | 1.43% |
| Chronic pancreatitis | 1660 | 21 | 1.27% | 996 | 5 | 0.50% |
| Severe kidney disease | 4982 | 259 | 5.20% | 3911 | 62 | 1.59% |
| Chronic kidney failure | 27517 | 1068 | 3.88% | 31472 | 394 | 1.25% |
| Diabetes mellitus | 74777 | 1050 | 1.40% | 88723 | 307 | 0.35% |
| Obesity | 90887 | 259 | 0.28% | 76384 | 118 | 0.15% |
| Cachexia/malnutrition | 1392 | 72 | 5.17% | 1388 | 51 | 3.67% |
| Coagulation disorder | 16820 | 891 | 5.30% | 14260 | 265 | 1.86% |
| Peritoneal adhesions*1 | 86353 | 604 | 0.70% | – | – | – |
| Acute cholecystitis, cholangitis, or biliary pancreatitis | 242894 | 2295 | 0.94% | – | – | – |
| Malignant neoplasia*2 | – | – | – | 12687 | 122 | 0.96% |
*1As the case definition for herniotomy excludes cases with adhesiolysis (because it excludes concomitant bowel surgery), peritoneal adhesions are not included as a secondary diagnosis.
*2The case definition for cholecystectomy excludes cases with malignant neoplasia as a secondary diagnosis
Acknowledgments
Translated from the original German by Caroline Shimakawa-Devitt, M.A.
Footnotes
Conflict of interest statement
The Department of Structural Advancement and Quality Management in Health Care, for which the authors work, is an endowed professorship of Helios Kliniken GmbH.
References
- 1.Nimptsch U, Mansky T. Krankheitsspezifische Versorgungsmerkmale in Deutschland: Analyse anhand der Bundesauswertung der German Inpatient Quality Indicators (G-IQI) Dtsch Med Wochenschr. 2012;137:1449–1457. doi: 10.1055/s-0032-1305086. [DOI] [PubMed] [Google Scholar]
- 2.Burgard G. Low Risk Operationen bei High Risk Patienten. In: Krahwinkel W, Meier-Hellmann A, Zacher J, editors. Peer Review. Sicher ist besser. Berlin: Medizinisch Wissenschaftliche Verlagsgesellschaft; 2013. pp. 49–56. [Google Scholar]
- 3.Frieling T. Cholecystektomie bei Leberzirrhose - eine Hochrisikokonstellation. In: Krahwinkel W, Meier-Hellmann A, Zacher J, editors. Peer Review. Sicher ist besser. Berlin: Medizinisch Wissenschaftliche Verlagsgesellschaft; 2013. pp. 145–147. [Google Scholar]
- 4.Forschungsdatenzentren der statistischen Ämter des Bundes und der Länder. Datenangebot | Fallpauschalenbezogene Krankenhausstatistik (DRG-Statistik) www.forschungsdatenzentren.de/bestand/drg/index.asp (last accessed on 15 May 2015) Wiesbaden: Forschungsdatenzentren der statistischen Ämter des Bundes und der Länder; [Google Scholar]
- 5.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Medical Care. 1998;36:8–27. doi: 10.1097/00005650-199801000-00004. [DOI] [PubMed] [Google Scholar]
- 6.AOK-Bundesverband, Forschungs- und Entwicklungsinstitut für das Sozial- und Gesundheitswesen Sachsen-Anhalt (FEISA), Helios Kliniken, Wissenschaftliches Institut der AOK (WIdO) Qualitätssicherung der stationären Versorgung mit Routinedaten (QSR) - Abschlussbericht. Bonn: WIdO; 2007. [Google Scholar]
- 7.Shah N, Hamilton M. Clinical review: Can we predict which patients are at risk of complications following surgery? Crit Care. 2013;17 doi: 10.1186/cc11904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Grade M, Quintel M, Ghadimi M. Standard perioperative management in gastrointestinal surgery. Langenbecks Arch Surg. 2011;396:591–606. doi: 10.1007/s00423-011-0782-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kuppinger D, Hartl WH, Bertok M, et al. Nutritional screening for risk prediction in patients scheduled for abdominal operations. Br J Surg. 2012;99:728–737. doi: 10.1002/bjs.8710. [DOI] [PubMed] [Google Scholar]
- 10.Hanley JA, Negassa A, Edwardes MD, Forrester JE. Statistical analysis of correlated data using generalized estimating equations: an orientation. Am J Epidemiol. 2003;157:364–375. doi: 10.1093/aje/kwf215. [DOI] [PubMed] [Google Scholar]
- 11.Cook NR. Use and misuse of the receiver operating characteristic curve in risk prediction. Circulation. 2007;115:928–935. doi: 10.1161/CIRCULATIONAHA.106.672402. [DOI] [PubMed] [Google Scholar]
- 12.LaValley MP. Statistical primer for cardiovascular research. Logistic regression. Circulation. 2008;117:2395–2399. doi: 10.1161/CIRCULATIONAHA.106.682658. [DOI] [PubMed] [Google Scholar]
- 13.Reppel M, Weil J. Einschränkung der Operationsfähigkeit. Was muss der Internist wissen? Internist. 2010;514:42–50. doi: 10.1007/s00108-009-2451-6. [DOI] [PubMed] [Google Scholar]
- 14.Böhmer AB, Wappler F, Zwißler B. Assessing preoperative risk—from routine tests to individualized investigation. Dtsch Arztebl Int. 2014;111:437–446. doi: 10.3238/arztebl.2014.0437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Deutsche Gesellschaft für Anästhesiologie und Intensivmedizin, Deutsche Gesellschaft für Innere Medizin, Deutsche Gesellschaft für Chirurgie. Präoperative Evaluation erwachsener Patienten vor elektiven, nichtkardiochirurgischen Eingriffen. Kardiologe. 2011;5:13–26. [Google Scholar]
- 16.Khatib-Chahidi K, Troja A, Kramer M, Klompmaker M, Raab HR, Antolovic D. Präoperatives Management bei Mangelernährten in der Viszeralchirurgie. Chirurg. 2014;85:520–528. doi: 10.1007/s00104-013-2637-2. [DOI] [PubMed] [Google Scholar]
- 17.Gustafsson UO, Ljungqvist O. Perioperative nutritional management in digestive tract surgery. Curr Opin Clin Nutr Metab Care. 2011;14:504–509. doi: 10.1097/MCO.0b013e3283499ae1. [DOI] [PubMed] [Google Scholar]
- 18.Upshaw J, Kiernan MS. Preoperative cardiac risk assessment for noncardiac surgery in patients with heart failure. Curr Heart Fail Rep. 2013;10:147–156. doi: 10.1007/s11897-013-0136-x. [DOI] [PubMed] [Google Scholar]
- 19.Ghaferi AA, Birkmeyer JD, Dimick JB. Variation in hospital mortality associated with inpatient surgery. N Engl J Med. 2009;361:1368–1375. doi: 10.1056/NEJMsa0903048. [DOI] [PubMed] [Google Scholar]
- 20.Gastmeier P, Brunkhorst F, Schrappe M, Kern W, Geffers C. Wie viele nosokomiale Infektionen sind vermeidbar? Dtsch Med Wochenschr. 2010;135:91–93. doi: 10.1055/s-0029-1244823. [DOI] [PubMed] [Google Scholar]
- 21.Devereaux PJ, Goldman L, Yusuf S, Gilbert K, Leslie K, Guyatt GH. Surveillance and prevention of major perioperative ischemic cardiac events in patients undergoing noncardiac surgery: a review. CMAJ. 2005;173:779–788. doi: 10.1503/cmaj.050316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Brienza N, Giglio MT, Marucci M. Preventing acute kidney injury after noncardiac surgery. Curr Opin Crit Care. 2010;16:353–358. doi: 10.1097/MCC.0b013e32833a9ef5. [DOI] [PubMed] [Google Scholar]
- 23.Hölscher UM, Gausmann P, Haindl H, et al. Patientensicherheit als nationales Gesundheitsziel: Status und notwendige Handlungsfelder für die Gesundheitsversorgung in Deutschland. Z Evid Fortbild Qual Gesundhwes. 2014;108:6–14. doi: 10.1016/j.zefq.2014.01.009. [DOI] [PubMed] [Google Scholar]
- 24.Haynes AB, Weiser TG, et al. Safe Surgery Saves Lives Study Group. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360:491–499. doi: 10.1056/NEJMsa0810119. [DOI] [PubMed] [Google Scholar]
- 25.Fudickar A, Hörle K, Wiltfang J, Bein B. The effect of the WHO Surgical Safety Checklist on complication rate and communication. Dtsch Arztebl Int. 2012;109:695–701. doi: 10.3238/arztebl.2012.0695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Richter A. Chirurgische Standards der perioperativen Patientenbehandlung. Chirurg. 2012;83:343–350. doi: 10.1007/s00104-011-2225-2. [DOI] [PubMed] [Google Scholar]
- 27.Murphy MM, Ng SC, Simons JP, Csikesz NG, Shah SA, Tseng JF. Predictors of major complications after laparoscopic cholecystectomy: surgeon, hospital, or patient? J Am Coll Surg. 2010;211:73–80. doi: 10.1016/j.jamcollsurg.2010.02.050. [DOI] [PubMed] [Google Scholar]
- 28.Harboe KM, Bardram L. Nationwide quality improvement of cholecystectomy: results from a national database. Int J Qual Health Care. 2011;23:565–573. doi: 10.1093/intqhc/mzr041. [DOI] [PubMed] [Google Scholar]
- 29.Helgstrand F, Rosenberg J, Kehlet H, Jorgensen LN, Bisgaard T. Nationwide prospective study of outcomes after elective incisional hernia repair. J Am Coll Surg. 2013;216:217–228. doi: 10.1016/j.jamcollsurg.2012.10.013. [DOI] [PubMed] [Google Scholar]
- 30.AQUA - Institut für angewandte Qualitätsförderung und Forschung im Gesundheitswesen GmbH. Qualitätsreport 2013. Göttingen: AQUA-Institut; 2014. 17 pp. [Google Scholar]