Morbidity and mortality in elderly patients undergoing evacuation of acute traumatic subdural hematoma

View More View Less
  • 1 Department of Neurologic Surgery, Mayo Clinic; and
  • 2 Mayo Clinic Neuro-Informatics Laboratory, Mayo Clinic, Rochester, Minnesota
Free access

OBJECTIVE

Acute traumatic subdural hematoma (atSDH) can be a life-threatening neurosurgical emergency that necessitates immediate evacuation. The elderly population can be particularly vulnerable to tearing bridging veins. The aim of this study was to evaluate inpatient morbidity and mortality, as well as predictors of inpatient mortality, in a national trauma database.

METHODS

The authors queried the 2016–2017 National Trauma Data Bank registry for patients aged 65 years and older who had undergone evacuation of atSDH. Patients were categorized into three age groups: 65–74, 75–84, and 85+ years. A multivariable logistic regression model was fitted for inpatient mortality adjusting for age group, sex, race, presenting Glasgow Coma Scale (GCS) category (3–8, 9–12, and 13–15), Injury Severity Score, presence of coagulopathy, presence of additional hemorrhages (epidural hematoma [EDH], intraparenchymal hematoma [IPH], and subarachnoid hemorrhage [SAH]), presence of midline shift > 5 mm, and pupillary reactivity (both, one, or none).

RESULTS

A total of 2508 patients (35% females) were analyzed. Age distribution was as follows: 990 patients at 65–74 years, 1096 at 75–84, and 422 at 85+. Midline shift > 5 mm was present in 72% of cases. With regard to additional hemorrhages, SAH was present in 21%, IPH in 10%, and EDH in 2%. Bilaterally reactive pupils were noted in 90% of patients. A major complication was observed in 14.4% of patients, and the overall mortality rate was 18.3%. In the multivariable analysis, the presenting GCS category was found to be the strongest predictor of postoperative inpatient mortality (3–8 vs 13–15: OR 3.63, 95% CI 2.68–4.92, p < 0.001; 9–12 vs 13–15: OR 2.64, 95% CI 1.79–3.90, p < 0.001; 30% of overall variation), followed by the presence of SAH (OR 2.86, 95% CI 2.21–3.70, p < 0.001; 25% of overall variation) and the presence of midline shift > 5 mm (OR 2.40, 95% CI 1.74–3.32, p < 0.001; 11% of overall variation). Model discrimination was excellent (c-index 0.81). Broken down by age decile group, mortality increased from 8.0% to 15.4% for GCS 13–15 to around 36% for GCS 9–12 to almost as high as 60% for GCS 3–8, particularly in those aged 85 years and older.

CONCLUSIONS

The present results from a national trauma database will, the authors hope, assist surgeons in preoperative discussions with patients and their families with regard to expected postoperative outcomes following surgical evacuation of an atSDH.

ABBREVIATIONS atSDH = acute traumatic SDH; EDH = epidural hematoma; GCS = Glasgow Coma Scale; ICP = intracranial pressure; IPH = intraparenchymal hematoma; ISS = Injury Severity Score; NSQIP = National Surgical Quality Improvement Program; NTDB = National Trauma Data Bank; SAH = subarachnoid hemorrhage; SDH = subdural hematoma; TBI = traumatic brain injury.

OBJECTIVE

Acute traumatic subdural hematoma (atSDH) can be a life-threatening neurosurgical emergency that necessitates immediate evacuation. The elderly population can be particularly vulnerable to tearing bridging veins. The aim of this study was to evaluate inpatient morbidity and mortality, as well as predictors of inpatient mortality, in a national trauma database.

METHODS

The authors queried the 2016–2017 National Trauma Data Bank registry for patients aged 65 years and older who had undergone evacuation of atSDH. Patients were categorized into three age groups: 65–74, 75–84, and 85+ years. A multivariable logistic regression model was fitted for inpatient mortality adjusting for age group, sex, race, presenting Glasgow Coma Scale (GCS) category (3–8, 9–12, and 13–15), Injury Severity Score, presence of coagulopathy, presence of additional hemorrhages (epidural hematoma [EDH], intraparenchymal hematoma [IPH], and subarachnoid hemorrhage [SAH]), presence of midline shift > 5 mm, and pupillary reactivity (both, one, or none).

RESULTS

A total of 2508 patients (35% females) were analyzed. Age distribution was as follows: 990 patients at 65–74 years, 1096 at 75–84, and 422 at 85+. Midline shift > 5 mm was present in 72% of cases. With regard to additional hemorrhages, SAH was present in 21%, IPH in 10%, and EDH in 2%. Bilaterally reactive pupils were noted in 90% of patients. A major complication was observed in 14.4% of patients, and the overall mortality rate was 18.3%. In the multivariable analysis, the presenting GCS category was found to be the strongest predictor of postoperative inpatient mortality (3–8 vs 13–15: OR 3.63, 95% CI 2.68–4.92, p < 0.001; 9–12 vs 13–15: OR 2.64, 95% CI 1.79–3.90, p < 0.001; 30% of overall variation), followed by the presence of SAH (OR 2.86, 95% CI 2.21–3.70, p < 0.001; 25% of overall variation) and the presence of midline shift > 5 mm (OR 2.40, 95% CI 1.74–3.32, p < 0.001; 11% of overall variation). Model discrimination was excellent (c-index 0.81). Broken down by age decile group, mortality increased from 8.0% to 15.4% for GCS 13–15 to around 36% for GCS 9–12 to almost as high as 60% for GCS 3–8, particularly in those aged 85 years and older.

CONCLUSIONS

The present results from a national trauma database will, the authors hope, assist surgeons in preoperative discussions with patients and their families with regard to expected postoperative outcomes following surgical evacuation of an atSDH.

ABBREVIATIONS atSDH = acute traumatic SDH; EDH = epidural hematoma; GCS = Glasgow Coma Scale; ICP = intracranial pressure; IPH = intraparenchymal hematoma; ISS = Injury Severity Score; NSQIP = National Surgical Quality Improvement Program; NTDB = National Trauma Data Bank; SAH = subarachnoid hemorrhage; SDH = subdural hematoma; TBI = traumatic brain injury.

The incidence of traumatic brain injury (TBI) has been increasing in the United States despite the progress made in decreasing motor vehicle accidents.1 This increase is attributed to a rising elderly population (defined as an age ≥ 65 years), who are especially susceptible to TBI secondary to falls.2 Based on data from the National Inpatient Sample, there were 950,132 TBI-related hospitalizations and 107,666 TBI-related deaths among adults aged ≥ 65 years from 2000 to 2010, with falls representing the most common mechanism of injury.3 More importantly, geriatric patients experience worse mortality rates and functional outcomes than nonelderly TBI patients even though their head and overall injury burdens are seemingly less severe.4 These patients also tend to have increased dependence postinjury given their decreased brain reserves.5

Elderly patients are particularly prone to shearing of cortical bridging veins even in the setting of minor trauma, predisposing them to acute subdural hematomas (SDHs). The frequent use of antithrombotic agents in this population may further exacerbate the risk of complications and death.6 Previous analyses of the National Inpatient Sample and National Surgical Quality Improvement Program (NSQIP) have demonstrated the overall risk of inpatient and 30-day mortality among patients with SDH (including both traumatic and nontraumatic) treated operatively to be 12% and 17%, respectively.7,8 An older age has been shown to be a predictor of adverse outcomes following surgical evacuation; however, the effect of surgical intervention on short- and longer-term functional outcomes has not been well characterized in the elderly.7,9 Currently available evidence on prognosis following acute traumatic SDH (atSDH) evacuation in this population consists of single-institution studies.10–17 Moreover, there is a paucity of literature elucidating the risk factors predictive of outcomes. To further complicate things, advance directives are not always available at the time of critical decision-making or are rather vague with regard to patient wishes; thus, families often find themselves in situations in which it is unclear whether surgery would serve the patient’s best interests and wishes. The financial and emotional costs to patients and their families are sometimes not taken into account, particularly amid an emergency department crisis when a lifesaving decision needs to be made within a few minutes.18,19 The availability of representative outcomes data could help facilitate discussions between providers and families to allow for more informed decision-making. Herein, we sought to determine the outcomes of surgical evacuation of atSDH in elderly patients, as well as elucidate factors predictive of outcomes, by utilizing a national trauma registry.

Methods

Data Source and Inclusion Criteria

We queried the National Trauma Data Bank (NTDB) for the years 2016–2017 for patients aged 65 years and older who had undergone craniotomy for a diagnosis of atSDH. To identify our cohort, we utilized the ICD-10 diagnosis codes denoting atSDH (Supplemental Table 1) in combination with an evacuation procedure code (00940ZZ, 00940ZX, 009400Z). The NTDB represents the largest aggregation of trauma registry data in the United States.20 It was created in 1989 by the American College of Surgeons Committee on Trauma to serve as a “central national repository of data on trauma and trauma care.”20 The registry contains a wealth of information, including demographics, comorbidities, injury severity, hospital characteristics, and inpatient outcomes, on over 8 million records from 765 hospitals across the country (2017 data).21 The data contained in the NTDB is de-identified and therefore exempt from institutional review board approval.

Outcomes of Interest

Inpatient mortality was the primary outcome of interest. Secondary outcomes consisted of inpatient major complication rate, ICU length of stay, and discharge disposition.

Covariates of Interest

We recorded the following covariates: demographics (sex, race, ethnicity, primary insurance payer, BMI), comorbidities and functional status, injury characteristics (vital signs upon presentation to the emergency department, Injury Severity Score [ISS], Glasgow Coma Scale [GCS] score, pupillary response, presence of brain midline shift > 5 mm [as defined in the National Trauma Data Standard dictionary]), and presence of additional intracranial hemorrhages (Supplemental Table 2), as well as complications in the postoperative setting.

Statistical Analysis

Descriptive statistics (medians with interquartile ranges for continuous data and frequencies with proportions for categorical data) are used to present available information. The Wilcoxon rank-sum test and Pearson’s chi-square test were used to compare continuous and categorical variables, respectively. Age was analyzed as a continuous variable and was stratified in the following groups in order to facilitate comparisons: 65–74 years, 75–84 years, and 85+ years. A multivariable logistic regression model was fitted for inpatient mortality adjusting for age, sex, race, GCS score at presentation, ISS, presence of bleeding disorder, presence of additional intracranial hemorrhages (epidural hematoma [EDH], intraparenchymal hematoma [IPH], and subarachnoid hemorrhage [SAH]), presence of midline shift, and pupillary response. To assess the robustness of our findings, we repeated the multivariable analysis within each age decile group.

Collinearity between the included covariates was examined using the variance inflation factor. Model discrimination was assessed using the c-statistic. Missing variables were imputed using multiple imputation from the rms package, which employs a combination of additive regression, bootstrapping, and predictive mean matching.22 A total of 10 imputed data sets were created, and imputation-specific coefficients were subsequently pooled to produce a single result. Finally, we also assessed the relative impact of each variable on inpatient mortality by using an importance metric defined as Wald chi-square penalized by the factor degrees of freedom, with a higher score indicating higher predictive importance.23 Statistical analysis was performed using R, an open-source software (R Foundation for Statistical Computing, https://www.R-project.org/), and the rms package. A p value was statistically significant if it was less than 0.05.

Results

Patient Demographics and Comorbidities

A total of 2508 patients (median age 77 years, 35% females) were included in the analysis. The majority of patients were White (1964 [80%]), while only 8% (190) were Hispanic. The most common primary insurance payer was Medicare (1926 [78%]), followed by private insurance (362 [14.6%]). Significant differences were observed among the age groups with regard to sex (p = 0.008), BMI (p < 0.001), insurance status (p < 0.001), and functional dependency (p = 0.001), as well as the presence of hypertension (p < 0.001), a bleeding disorder (p = 0.04), dementia (< 0.001), a drug use disorder (p < 0.001), and alcohol abuse (p < 0.001). This information is summarized in Table 1.

TABLE 1.

Summary of patient characteristics and comorbidities per age group

VariableAvailable Data (no. of cases)All (n = 2508)65–74 Yrs (n = 990)75–84 Yrs (n = 1096)85+ Yrs (n = 422)p Value
Female sex, no. (%)2508873 (34.8)311 (31.4)396 (36.1)166 (39.3)0.008
Race, no. (%)24630.19
 White1964 (79.7)750 (77.6)870 (80.4)344 (83.1)
 Black227 (9.2)106 (11.0)90 (8.3)31 (7.5)
 Asian102 (4.1)42 (4.3)48 (4.4)12 (2.9)
 Other170 (6.9)69 (7.1)74 (6.8)27 (6.5)
Hispanic ethnicity, no. (%)2386190 (8.0)81 (8.6)81 (7.8)28 (6.9)0.55
Median BMI [IQR]250825.8 [22.8, 29.3]26.6 [23.1, 30.4]25.8 [23.0, 29.0]24.7 [21.9, 27.4]<0.001
Insurance status, no. (%)2472<0.001
 Private362 (14.6)179 (18.4)132 (12.2)51 (12.2)
 Medicare1926 (77.9)690 (70.8)883 (81.8)353 (84.2)
 Medicaid70 (2.8)36 (3.7)30 (2.8)4 (1.0)
 Self-pay42 (1.7)26 (2.7)12 (1.1)4 (1.0)
 Other72 (2.9)43 (4.4)22 (2.0)7 (1.7)
Dependent functional status, no. (%)2508338 (13.5)109 (11.0)152 (13.9)77 (18.2)0.001
Comorbidities, no. (%)2508
 Hypertension1682 (67.1)621 (62.7)755 (68.9)306 (72.5)<0.001
 On chronic steroids26 (1.0)13 (1.3)8 (0.7)5 (1.2)0.38
 Cirrhosis24 (1.0)12 (1.2)10 (0.9)2 (0.5)0.44
 Alcohol abuse154 (6.1)107 (10.8)37 (3.4)10 (2.4)<0.001
 Coagulopathy366 (14.6)123 (12.4)179 (16.3)64 (15.2)0.04
 Dementia279 (11.1)66 (6.7)147 (13.4)66 (15.6)<0.001
 Chronic renal failure96 (3.8)39 (3.9)46 (4.2)11 (2.6)0.34
 On chemotherapy17 (0.7)11 (1.1)4 (0.4)2 (0.5)0.11
 Drug use disorder32 (1.3)27 (2.7)4 (0.4)1 (0.2)<0.001

Boldface type indicates statistical significance.

Baseline Injury Severity Characteristics

The median ISS was 25 (IQR 17–26). The most common presenting GCS score was 13–15 (1722 [73%]), followed by 3–8 (449 [19.1%]) and 9–12 (177 [7.5%]). With regard to pupillary response, 90% of the patients (n = 2053) had both reactive pupils, whereas 3.6% had one reactive pupil and 6.4% had nonreactive pupils. Midline shift greater than 5 mm was observed in 72% of the cohort (n = 1676). In terms of additional hemorrhages, an EDH was encountered in 2% of the patients, SAH in 21%, and IPH in 10%. We observed significant differences among the groups with regard to pupillary response (overall p = 0.004), as well as the presence of EDH (p = 0.046), SAH (p < 0.001), and IPH (p = 0.001). Finally, an external ventricular drain and intraparenchymal monitor were placed in 20% and 6% of patients, respectively. Patients aged 85+ years were significantly less likely to receive an intracranial pressure (ICP) monitoring device than those who were 75–84 and those who were 65–74 (20.6% vs 23% vs 28%, respectively; p = 0.004). This information is presented in Table 2.

TABLE 2.

Summary of baseline injury characteristics

VariableAvailable Data (no. of cases)All (n = 2508)65–74 Yrs (n = 990)75–84 Yrs (n = 1096)85+ Yrs (n = 422)p Value
Vital signs in ED, median [IQR]2508
 Systolic blood pressure151 [134, 170]150 [133, 169]151 [134, 171]153 [136, 172]0.049
 Heart rate80.0 [69.0, 92.0]81.0 [70.0, 95.0]80.0 [68.0, 91.0]78.0 [69.0, 89.0]<0.001
 Respiratory rate18.0 [16.0, 20.0]18.0 [16.0, 20.0]18.0 [16.0, 20.0]18.0 [16.0, 20.0]0.85
 Oxygen saturation98.0 [96.0, 99.0]98.0 [96.0, 100]97.0 [96.0, 99.0]97.0 [95.0, 99.0]<0.001
Median ISS [IQR]250825.0 [17.0, 26.0]25.0 [18.0, 26.0]25.0 [17.0, 26.0]25.0 [17.0, 26.0]0.06
Total GCS score, no. (%)2348<0.001
 3–8449 (19.1)227 (24.4)175 (16.9)47 (12.2)
 9–12177 (7.5)78 (8.4)69 (6.7)30 (7.8)
 13–151722 (73.3)625 (67.2)789 (76.4)308 (80.0)
Pupillary response, no. (%)22820.004
 Both reactive2053 (90.0)798 (87.5)904 (90.7)351 (94.1)
 One reactive83 (3.6)41 (4.5)37 (3.7)5 (1.3)
 Neither reactive146 (6.4)73 (8.0)56 (5.6)17 (4.6)
MLS >5 mm, no. (%)23401676 (71.6)687 (74.0)723 (70.4)266 (68.6)0.08
Other injuries, no. (%)2508
 EDH57 (2.3)31 (3.1)21 (1.9)5 (1.2)0.046
 SAH534 (21.3)259 (26.2)197 (18.0)78 (18.5)<0.001
 IPH247 (9.8)124 (12.5)93 (8.5)30 (7.1)0.001
ICP monitor, no. (%)2508615 (24.5)276 (27.9)252 (23.0)87 (20.6)0.004
EVD, no. (%)2508494 (19.7)205 (20.7)210 (19.2)79 (18.7)
Intraparenchymal catheter, no. (%)2508142 (5.7)79 (8.0)53 (4.8)10 (2.4)
Licox monitor, no. (%)250814 (0.6)7 (0.7)4 (0.4)3 (0.7)>0.99
Jugular venous bulb monitor, no. (%)250811 (0.4)8 (0.8)3 (0.3)0 (0)>0.99

ED = emergency department; EVD = external ventricular drain; MLS = midline shift.

Boldface type indicates statistical significance.

Postoperative Outcomes

At least one major complication occurred in 14.4% of patients with no significant difference with regard to age group (p = 0.15). Unplanned return to the operating room was the most common complication (4.9%), followed by cerebrovascular accident (2.1%) and cardiac arrest (1.4%). There was no major difference among the three groups in terms of complications (p = 0.15). Complication and outcome information is summarized in Tables 3 and 4.

TABLE 3.

Summary of postoperative complications

Major ComplicationAll (n = 2508)65–74 Yrs (n = 990)75–84 Yrs (n = 1096)85+ Yrs (n = 422)p Value
Total362 (14.4)158 (16.0)142 (13.0)62 (14.7)0.15
 ARDS27 (1.08)16 (1.62)7 (0.64)4 (0.95)0.10
 Cardiac arrest36 (1.44)20 (2.02)13 (1.19)3 (0.71)0.11
 Myocardial infarction12 (0.48)5 (0.51)4 (0.36)3 (0.71)0.59
 Pulmonary embolism19 (0.76)9 (0.91)8 (0.73)2 (0.47)0.79
 Cerebrovascular accident53 (2.11)21 (2.12)27 (2.46)5 (1.18)0.30
 Sepsis18 (0.72)5 (0.51)8 (0.73)5 (1.18)0.36
 Catheter-related bloodstream infection2 (0.08)2 (0.20)0 (0.00)0 (0.00)0.46
 Ventilator-associated pneumonia58 (2.31)36 (3.64)17 (1.55)5 (1.18)0.002
 Unplanned return to OR122 (4.86)61 (6.16)38 (3.47)23 (5.45)0.01

ARDS = acute respiratory distress syndrome; OR = operating room.

Only the most common major complications are listed. Values are expressed as the number (%) of patients, unless indicated otherwise. Boldface type indicates statistical significance.

TABLE 4.

Summary of postoperative outcomes

VariableAvailable Data (no. of cases)All (n = 2508)65–74 Yrs (n = 990)75–84 Yrs (n = 1096)85+ Yrs (n = 422)p Value
ICU LOS, median [IQR]25086.00 [4.00, 10.0]6.00 [4.00, 11.0]5.00 [3.00, 9.00]5.00 [4.00, 9.00]0.001
Ventilation days, median [IQR]25085.00 [2.00, 10.0]6.00 [2.00, 11.0]5.00 [2.00, 9.00]4.00 [2.00, 9.00]0.037
Inpatient mortality, no. (%)2508460 (18.3)172 (17.4)199 (18.2)89 (21.1)0.24
Withdrawal of care, no. (%)2392420 (17.6)147 (15.5)180 (17.2)93 (23.3)0.003
Discharge disposition, no. (%)2508<0.001
 Home/home health469 (18.7)228 (23.0)202 (18.4)39 (9.2)
 Nursing home546 (21.8)150 (15.2)266 (24.3)130 (30.8)
 Inpatient rehab647 (25.8)273 (27.6)274 (25.0)100 (23.7)
 Short-term inpatient care37 (1.5)23 (2.3)12 (1.1)2 (0.5)
 Hospice184 (7.3)56 (5.7)86 (7.8)42 (10.0)
 Died460 (18.3)172 (17.4)199 (18.2)89 (21.1)
 Other51 (2.03)25 (2.53)20 (1.82)6 (1.42)

LOS = length of stay.

Boldface type indicates statistical significance.

Inpatient Mortality

Patients who died during their hospital stay were more likely to have a coagulopathy (including antithrombotic use; 19% vs 14%, p = 0.01), cirrhosis (2% vs 0.7%, p = 0.03), and chronic renal failure (5.6% vs 3.4%, p = 0.035). These patients also had lower total GCS scores (10 vs 15, p < 0.001) and were more likely to have one reactive pupil (7.1% vs 2.8%) or no reactive pupils (18% vs 3.7%, overall p < 0.001). Finally, the presence of EDH (4.6% vs 1.8%, p = 0.001), SAH (46% vs 16%, p < 0.001), or IPH (21% vs 7.4%, p < 0.001) was also more frequent in patients with inpatient mortality. The results of univariate analyses are presented in Supplemental Tables 3–5.

Multivariable Analysis

In the multivariable logistic regression analysis, age (75–84 vs 65–74: OR 1.57, 95% CI 1.21–2.04, p < 0.001; 85+ vs 65–74: OR 2.26, 95% CI 1.61–3.15, p < 0.001), lower total GCS score (3–8 vs 13–15: OR 3.63, 95% CI 2.68–4.92, p < 0.001; 9–12 vs 13–15: OR 2.64, 95% CI 1.79–3.90, p < 0.001), higher ISS (OR 1.56, 95% CI 1.32–1.85, p < 0.001), presence of SAH (OR 2.86, 95% CI 2.21–3.70, p < 0.001), presence of IPH (OR 1.68, 95% CI 1.20–2.34, p = 0.02), and coagulopathy/antithrombotic use (OR 1.41, 95% CI 1.05–1.91, p = 0.02) were associated with increased inpatient mortality. Patients with no reactive pupils also had significantly higher odds of death during their hospital stay (OR 1.81, 95% CI 1.18–2.76, p = 0.006). Model discrimination was found to be excellent (c-index 0.81). The results of the multivariable analysis are presented in Table 5.

TABLE 5.

Results of multivariable logistic regression analysis for inpatient mortality

VariableOR95% CIp Value
Age 75–84 vs 65–74 yrs1.571.21–2.04<0.001
Age 85+ vs 65–74 yrs2.261.61–3.15<0.001
F vs M sex0.860.67–1.090.21
Black vs White race0.580.35–0.950.03
Asian vs White race0.690.36–1.340.28
Other vs White race0.750.47–1.210.24
Total GCS score 3–8 vs 13–153.632.68–4.92<0.001
Total GCS score 9–12 vs 13–152.641.79–3.90<0.001
ISS (26:17)*1.561.32–1.85<0.001
Coagulopathy1.411.05–1.910.02
Presence of IPH1.681.20–2.340.02
Presence of EDH1.860.99–3.480.052
Presence of SAH2.862.21–3.70<0.001
Presence of MLS >5 mm2.401.74–3.320.001
One vs both reactive pupils1.360.79–2.340.27
No vs both reactive pupils1.811.18–2.760.006

Boldface type indicates statistical significance. The c-index was 0.81 (excellent discrimination).

Odds ratios for continuous variables are presented as 75th:25th percentile.

Predictor importance is graphically depicted in Fig. 1. The three most important predictors of inpatient mortality were, in order of decreasing frequency, presenting GCS category (30% of overall Wald chi-square), presence of traumatic SAH (25% of overall Wald chi-square), and presence of midline shift > 5 mm (11% of overall Wald chi-square).

FIG. 1.
FIG. 1.

Plot of predictor variables in decreasing importance. MLS = midline shift.

Subgroup Analysis

When we subsetted our sample based on age decile, we found the same factors to be significant, that is, presenting GCS score, midline shift greater than 5 mm, presence of SAH, and ISS. These results are presented in Supplemental Tables 6–8. We also graphed mortality rates within each age decile group (Fig. 2). Mortality rates ranged from 8.0% to 15.4% for GCS 13–15, 27.6%–36.2% for GCS 9–12, and 41.9%–57.4% for GCS 3–8.

FIG. 2.
FIG. 2.

Inpatient mortality rates broken down by age group and presenting GCS score. All within-group comparisons were statistically significant (p < 0.001).

Discussion

In the present study, we investigated inpatient outcomes following surgical evacuation of an atSDH in a cohort of 2508 elderly patients from a national trauma registry. In summary, we found an overall inpatient mortality rate of 18.3%. Within the group of patients presenting with GCS 3–8, inpatient mortality was found to be 42%, 46%, and 57% for the 65–74, 75–84, and 85+ age groups, respectively. In addition, the three most important independent predictors of inpatient mortality were presenting GCS score, the presence of traumatic SAH, and the presence of midline shift (> 5 mm according to trauma criteria).

The current literature consists of small, single-institution series and has yielded variable results. According to Evans and colleagues’ recently published systematic review of 7 studies and 396 elderly patients with a diagnosis of atSDH (sample size range 27–119), 30-day mortality ranged from 27% to 70%, most likely reflecting disparities in management, baseline population mix, and injury severity.17 Interestingly, the authors found an initial GCS score of 3–8 to be the most consistent negative prognostic factor, an observation that corroborates the results of our risk-adjusted analysis. It should be noted though that both operative and nonoperative cases were included in that review, which could confound the interpretation of results. Finally, in an NSQIP 2005–2012 study of 746 craniotomy cases for SDH, 30-day mortality was 17%.7 A higher American Society of Anesthesiologists (ASA) class, ascites, a “bleeding disorder,” and coma correlated with higher odds of 30-day mortality.

An important finding in our analysis was the association between antithrombotic use/coagulopathy and postoperative mortality. In a study of 157 elderly patients, Rozzelle et al. found that a GCS score of 7 or less and an age greater than 80 correlated with worse inpatient mortality; previous antithrombotic medication use, nature of the trauma, comorbidity score, and presence of midline shift were not found to be significant.24 On the contrary, Scotti et al. showed that anticoagulation use (warfarin or novel oral anticoagulants), as well as dual antiplatelet therapy, in elderly TBI patients was associated with a higher mortality risk and functional dependency at discharge.25 In a similar fashion, Baraniskin et al. observed inferior outcomes in acute SDH patients on prehospital antithrombotic therapy (OR for recurrence or in-hospital mortality: 2.86–5.50, all p < 0.05).6

Although pupillary response was found to be a statistically significant and important negative predictor for in-hospital mortality in our analysis, others have not found a significant relationship.10,13–16 Pupillary changes can occur from uncal herniation due to mass effect leading to compression of the oculomotor nerve and brainstem, but they can also occur due to orbital/ocular trauma; therefore, the true etiology of the pupillary changes must be understood before it can be used in treatment or prognostic decisions. In a systematic review by Evans et al., the strength of currently available evidence for pupillary response as a prognostic factor was “very low” because of inconsistencies and limitations in the study design.17 It is interesting to note that pupillary response was nonsignificant in studies that included very elderly patients (age ≥ 75 years and ≥ 80 years), as opposed to the studies that defined the elderly as those older than 65 years.10,15 This might be due to the fact that pupillary reaction may already be highly sluggish at baseline in very elderly patients, thereby leading to questionable prognostic value in this subpopulation. It is also important to note that midline shift greater than 5 mm was also found to be independently associated with mortality in our analysis, with relatively high predictive importance. Petridis et al. further defined the relationship between mass effect from the hematoma causing midline shift and found that midline shift greater than hematoma width was associated with adverse outcomes (on univariate analysis alone).14 Midline shift should likely be evaluated on a continuous variable basis, as each increased millimeter likely increases morbidity and mortality. In a secondary analysis of the Citicoline Brain Injury Treatment (COBRIT) trial, Puffer et al. demonstrated worsening outcomes as midline shift increased from no shift to 1–5 mm, further worsening at shifts of 6–10 mm, and significantly poor outcomes with shifts > 10 mm, as expected.26 However, the trial was inclusive of all TBI patients and not limited to elderly patients undergoing surgical evacuation of atSDH. Nonetheless, increasing midline shift at presentation correlates with worse outcomes, but like all individual patient characteristics, this imaging variable should not be used alone as a prognostic factor but should be interpreted with the overall injury severity and comorbidity burden of each patient encountered.

Prior evidence exploring the relationship among ICP monitoring, older age, and TBI outcomes is very limited. Interestingly, a study by Thompson et al. showed that older TBI patients are offered a lower intensity of care after adjusting for sex, comorbidities, and injury-related factors.27 In a similar fashion, Schupper et al. found age disparities with regard to adherence to the Brain Trauma Foundation guidelines among blunt TBI patients presenting with a GCS score of 3–8.28 The differential effect of ICP monitoring and age on short- and long-term outcomes following severe TBI certainly merits further investigation and remains to be determined in future research. Furthermore, whether ICP monitoring contributes to lower mortality rates in severe TBI remains an area of contention among the scientific community, with the literature demonstrating conflicting results. In our study, we decided not to include ICP monitoring in the risk-adjusted analysis, as we focused on factors that are available to surgeons at the time of initial evaluation.

Ultimately, early and accurate prognostication in TBI is often premature and very challenging; neurosurgeons are very commonly called upon to provide information regarding prognosis and what kind of goals can or cannot be achieved with each intervention. Advance directives can certainly be helpful to provide a guiding framework. A recent report demonstrated that an estimated 67% of seriously ill patients in 2010 had an advance directive compared to only 21% in 1994, primarily involving the elderly population.29 Despite their increasing use though, the utility of advance directives to document patient wishes remains uncertain.30 More importantly, a readily available, easily interpreted, and applicable advance directive may not be present in the acuity of a traumatic event; thus, a patient may have undesired, invasive procedures performed.30 It is of paramount importance to establish well-defined goals of care and set realistic expectations with the patient’s family, which is one of the most critical aspects of trauma surgery, including neurosurgery.

Study Limitations

Our study has several limitations. First, although it is registry based, there is still a risk for coding errors. Second, we were not able to discern with high specificity whether the procedure performed was craniotomy versus burr hole evacuation. Third, the large number of patients in database studies often leads to statistically significant differences in variables that may not be clinically relevant. Fourth, we were only able to investigate inpatient outcomes and not longer-term outcomes, including functionality and degree of independence. Fifth, we did not have access to laboratory parameters (e.g., platelet count) or specific antithrombotic medications (e.g., aspirin vs warfarin vs dabigatran, etc.) in order to decipher their role on postoperative mortality. Sixth, we could not analyze midline shift as a continuous variable. There may also be a risk of residual confounding, as some radiological and clinical covariates such as hematoma size and postoperative seizures were not available. This may be important, as seizures following evacuation have also been found to be associated with worse early functional outcomes, although these patients can still attain a delayed favorable recovery.11,31 Last, and most importantly, selection bias is always a concern when investigating outcomes in the elderly population; sicker, dependent patients are less likely to be offered an aggressive surgical intervention, unless the patient and/or their family decide to pursue every possible means of care regardless of the potential postoperative outcome. Although we were not able to measure this bias, our study essentially represents an actuarial analysis of elderly patients with atSDH in the United States on a national scale while preserving a degree of data granularity close to that seen with single-institution series. Furthermore, the large sample size allowed us to perform subgroup analysis as well as risk-adjusted analysis controlling for a multitude of clinically relevant confounding variables. Despite the level of detail, the registry lacks variables pertinent to estimating frailty in the TBI population, such as the level of independence in activities of daily living, nutritional status, degree of dementia when present, etc. Existing literature on this topic is rather limited; therefore, future research into the relationship between frailty indices (novel or previously established) and short- and long-term outcomes is highly encouraged.

Conclusions

In summary, the findings of our nationwide study suggest an overall in-hospital mortality rate of 18% and a major complication rate of 14.4% among elderly patients (age ≥ 65) undergoing surgical evacuation of atSDH. In a multivariable analysis, the three most important predictors were found to be the presenting GCS score, presence of traumatic SAH, and presence of midline shift > 5 mm. The same factors were also significant in a subgroup analysis by age decile. Finally, we showed that mortality increased from 8% to 15.4% for patients with a GCS 13–15 to around 36% for GCS 9–12, and almost as high as 60% for GCS 3–8, particularly in those aged 85 and older.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Bydon, Kerezoudis, Goyal. Acquisition of data: Bydon, Kerezoudis, Goyal. Analysis and interpretation of data: Kerezoudis, Goyal, Puffer, Parney, Meyer. Drafting the article: Kerezoudis, Goyal, Puffer, Parney, Meyer. Critically revising the article: Kerezoudis, Goyal, Puffer, Parney, Meyer. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Bydon.

Supplemental Information

Online-Only Content

Supplemental material is available online.

References

  • 1

    TBI Data and Statistics. Centers for Disease Control and Prevention. Accessed August 20, 2020. https://www.cdc.gov/traumaticbraininjury/data/index.html

    • Export Citation
  • 2

    Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic brain injury-related emergency department visits, hospitalizations, and deaths—United States, 2007 and 2013. MMWR Surveill Summ. 2017;66(9):116.

    • Search Google Scholar
    • Export Citation
  • 3

    Haring RS, Narang K, Canner JK, Traumatic brain injury in the elderly: morbidity and mortality trends and risk factors. J Surg Res. 2015;195(1):19.

    • Search Google Scholar
    • Export Citation
  • 4

    Susman M, DiRusso SM, Sullivan T, Traumatic brain injury in the elderly: increased mortality and worse functional outcome at discharge despite lower injury severity. J Trauma. 2002;53(2):219224.

    • Search Google Scholar
    • Export Citation
  • 5

    Testa JA, Malec JF, Moessner AM, Brown AW. Outcome after traumatic brain injury: effects of aging on recovery. Arch Phys Med Rehabil. 2005;86(9):18151823.

    • Search Google Scholar
    • Export Citation
  • 6

    Baraniskin A, Steffens C, Harders A, Impact of pre-hospital antithrombotic medication on the outcome of chronic and acute subdural hematoma. J Neurol Surg A Cent Eur Neurosurg. 2014;75(1):3136.

    • Search Google Scholar
    • Export Citation
  • 7

    Lukasiewicz AM, Grant RA, Basques BA, Patient factors associated with 30-day morbidity, mortality, and length of stay after surgery for subdural hematoma: a study of the American College of Surgeons National Surgical Quality Improvement Program. J Neurosurg. 2016;124(3):760766.

    • Search Google Scholar
    • Export Citation
  • 8

    Frontera JA, Egorova N, Moskowitz AJ. National trend in prevalence, cost, and discharge disposition after subdural hematoma from 1998–2007. Crit Care Med. 2011;39(7):16191625.

    • Search Google Scholar
    • Export Citation
  • 9

    Mak CHK, Wong SKH, Wong GK, Traumatic brain injury in the elderly: is it as bad as we think? Curr Transl Geriatr Exp Gerontol Rep. 2012;1(3):171178.

    • Search Google Scholar
    • Export Citation
  • 10

    McGinity MJ, Michalek JE, Rodriguez JS, Floyd JR. Surgical evacuation of acute subdural hematoma in octogenarians: a ten-year experience from a single trauma center. Br J Neurosurg. 2017;31(6):714717.

    • Search Google Scholar
    • Export Citation
  • 11

    Won S-Y, Dubinski D, Brawanski N, Significant increase in acute subdural hematoma in octo- and nonagenarians: surgical treatment, functional outcome, and predictors in this patient cohort. Neurosurg Focus. 2017;43(5):E10.

    • Search Google Scholar
    • Export Citation
  • 12

    Benedetto N, Gambacciani C, Montemurro N, Surgical management of acute subdural haematomas in elderly: report of a single center experience. Br J Neurosurg. 2017;31(2):244248.

    • Search Google Scholar
    • Export Citation
  • 13

    Jamjoom A. Justification for evacuating acute subdural haematomas in patients above the age of 75 years. Injury. 1992;23(8):518520.

  • 14

    Petridis AK, Dörner L, Doukas A, Acute subdural hematoma in the elderly; clinical and CT factors influencing the surgical treatment decision. Cent Eur Neurosurg. 2009;70(2):7378.

    • Search Google Scholar
    • Export Citation
  • 15

    Raj R, Mikkonen ED, Kivisaari R, Mortality in elderly patients operated for an acute subdural hematoma: a surgical case series. World Neurosurg. 2016;88:592597.

    • Search Google Scholar
    • Export Citation
  • 16

    Taussky P, Hidalgo ET, Landolt H, Fandino J. Age and salvageability: analysis of outcome of patients older than 65 years undergoing craniotomy for acute traumatic subdural hematoma. World Neurosurg. 2012;78(3-4):306311.

    • Search Google Scholar
    • Export Citation
  • 17

    Evans LR, Jones J, Lee HQ, Prognosis of acute subdural hematoma in the elderly: a systematic review. J Neurotrauma. 2019;36(4):517522.

    • Search Google Scholar
    • Export Citation
  • 18

    Watanabe Y, Shiel A, McLellan DL, The impact of traumatic brain injury on family members living with patients: a preliminary study in Japan and the UK. Disabil Rehabil. 2001;23(9):370378.

    • Search Google Scholar
    • Export Citation
  • 19

    Duff D. Family impact and influence following severe traumatic brain injury. Axone. 2006;27(2):923.

  • 20

    Annual call for data: National Trauma Data Bank (NTDB). American College of Surgeons. Accessed August 20, 2020. https://www.facs.org/quality-programs/trauma/tqp/center-programs/ntd

    • Export Citation
  • 21

    Fantus RJ. Annual Report 2017: ICD-10. Bulletin. January 6, 2018. Accessed August 20, 2020. https://bulletin.facs.org/2018/01/annual-report-2017-icd-10/

    • Search Google Scholar
    • Export Citation
  • 22

    Harrell F. Regression modeling strategies. R package version 4.4-2. 2016. Accessed August 20, 2020. https://cran.r-project.org/web/packages/rms/index.html

    • Search Google Scholar
    • Export Citation
  • 23

    Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis. Springer Science & Business Media; 2013.

    • Search Google Scholar
    • Export Citation
  • 24

    Rozzelle CJ, Wofford JL, Branch CL. Predictors of hospital mortality in older patients with subdural hematoma. J Am Geriatr Soc. 1995;43(3):240244.

    • Search Google Scholar
    • Export Citation
  • 25

    Scotti P, Séguin C, Lo BWY, Antithrombotic agents and traumatic brain injury in the elderly population: hemorrhage patterns and outcomes. J Neurosurg. 2020;133(2):486495.

    • Search Google Scholar
    • Export Citation
  • 26

    Puffer RC, Yue JK, Mesley M, Long-term outcome in traumatic brain injury patients with midline shift: a secondary analysis of the Phase 3 COBRIT clinical trial. J Neurosurg. 2018;131(2):596603.

    • Search Google Scholar
    • Export Citation
  • 27

    Thompson HJ, Rivara FP, Jurkovich GJ, Evaluation of the effect of intensity of care on mortality after traumatic brain injury. Crit Care Med. 2008;36(1):282290.

    • Search Google Scholar
    • Export Citation
  • 28

    Schupper AJ, Berndtson AE, Smith A, Respect your elders: effects of ageing on intracranial pressure monitor use in traumatic brain injury. Trauma Surg Acute Care Open. 2019;4(1):e000306.

    • Search Google Scholar
    • Export Citation
  • 29

    Silveira MJ, Kim SYH, Langa KM. Advance directives and outcomes of surrogate decision making before death. N Engl J Med. 2010;362(13):12111218.

    • Search Google Scholar
    • Export Citation
  • 30

    Gordy S, Klein E. Advance directives in the trauma intensive care unit: do they really matter? Int J Crit Illn Inj Sci. 2011;1(2):132137.

    • Search Google Scholar
    • Export Citation
  • 31

    Rabinstein AA, Chung SY, Rudzinski LA, Lanzino G. Seizures after evacuation of subdural hematomas: incidence, risk factors, and functional impact. J Neurosurg. 2010;112(2):455460.

    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Supplementary Materials

Contributor Notes

Correspondence Mohamad Bydon: Mayo Clinic, Rochester, MN. bydon.mohamad@mayo.edu.

INCLUDE WHEN CITING DOI: 10.3171/2020.7.FOCUS20439.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

  • View in gallery

    Plot of predictor variables in decreasing importance. MLS = midline shift.

  • View in gallery

    Inpatient mortality rates broken down by age group and presenting GCS score. All within-group comparisons were statistically significant (p < 0.001).

  • 1

    TBI Data and Statistics. Centers for Disease Control and Prevention. Accessed August 20, 2020. https://www.cdc.gov/traumaticbraininjury/data/index.html

    • Export Citation
  • 2

    Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic brain injury-related emergency department visits, hospitalizations, and deaths—United States, 2007 and 2013. MMWR Surveill Summ. 2017;66(9):116.

    • Search Google Scholar
    • Export Citation
  • 3

    Haring RS, Narang K, Canner JK, Traumatic brain injury in the elderly: morbidity and mortality trends and risk factors. J Surg Res. 2015;195(1):19.

    • Search Google Scholar
    • Export Citation
  • 4

    Susman M, DiRusso SM, Sullivan T, Traumatic brain injury in the elderly: increased mortality and worse functional outcome at discharge despite lower injury severity. J Trauma. 2002;53(2):219224.

    • Search Google Scholar
    • Export Citation
  • 5

    Testa JA, Malec JF, Moessner AM, Brown AW. Outcome after traumatic brain injury: effects of aging on recovery. Arch Phys Med Rehabil. 2005;86(9):18151823.

    • Search Google Scholar
    • Export Citation
  • 6

    Baraniskin A, Steffens C, Harders A, Impact of pre-hospital antithrombotic medication on the outcome of chronic and acute subdural hematoma. J Neurol Surg A Cent Eur Neurosurg. 2014;75(1):3136.

    • Search Google Scholar
    • Export Citation
  • 7

    Lukasiewicz AM, Grant RA, Basques BA, Patient factors associated with 30-day morbidity, mortality, and length of stay after surgery for subdural hematoma: a study of the American College of Surgeons National Surgical Quality Improvement Program. J Neurosurg. 2016;124(3):760766.

    • Search Google Scholar
    • Export Citation
  • 8

    Frontera JA, Egorova N, Moskowitz AJ. National trend in prevalence, cost, and discharge disposition after subdural hematoma from 1998–2007. Crit Care Med. 2011;39(7):16191625.

    • Search Google Scholar
    • Export Citation
  • 9

    Mak CHK, Wong SKH, Wong GK, Traumatic brain injury in the elderly: is it as bad as we think? Curr Transl Geriatr Exp Gerontol Rep. 2012;1(3):171178.

    • Search Google Scholar
    • Export Citation
  • 10

    McGinity MJ, Michalek JE, Rodriguez JS, Floyd JR. Surgical evacuation of acute subdural hematoma in octogenarians: a ten-year experience from a single trauma center. Br J Neurosurg. 2017;31(6):714717.

    • Search Google Scholar
    • Export Citation
  • 11

    Won S-Y, Dubinski D, Brawanski N, Significant increase in acute subdural hematoma in octo- and nonagenarians: surgical treatment, functional outcome, and predictors in this patient cohort. Neurosurg Focus. 2017;43(5):E10.

    • Search Google Scholar
    • Export Citation
  • 12

    Benedetto N, Gambacciani C, Montemurro N, Surgical management of acute subdural haematomas in elderly: report of a single center experience. Br J Neurosurg. 2017;31(2):244248.

    • Search Google Scholar
    • Export Citation
  • 13

    Jamjoom A. Justification for evacuating acute subdural haematomas in patients above the age of 75 years. Injury. 1992;23(8):518520.

  • 14

    Petridis AK, Dörner L, Doukas A, Acute subdural hematoma in the elderly; clinical and CT factors influencing the surgical treatment decision. Cent Eur Neurosurg. 2009;70(2):7378.

    • Search Google Scholar
    • Export Citation
  • 15

    Raj R, Mikkonen ED, Kivisaari R, Mortality in elderly patients operated for an acute subdural hematoma: a surgical case series. World Neurosurg. 2016;88:592597.

    • Search Google Scholar
    • Export Citation
  • 16

    Taussky P, Hidalgo ET, Landolt H, Fandino J. Age and salvageability: analysis of outcome of patients older than 65 years undergoing craniotomy for acute traumatic subdural hematoma. World Neurosurg. 2012;78(3-4):306311.

    • Search Google Scholar
    • Export Citation
  • 17

    Evans LR, Jones J, Lee HQ, Prognosis of acute subdural hematoma in the elderly: a systematic review. J Neurotrauma. 2019;36(4):517522.

    • Search Google Scholar
    • Export Citation
  • 18

    Watanabe Y, Shiel A, McLellan DL, The impact of traumatic brain injury on family members living with patients: a preliminary study in Japan and the UK. Disabil Rehabil. 2001;23(9):370378.

    • Search Google Scholar
    • Export Citation
  • 19

    Duff D. Family impact and influence following severe traumatic brain injury. Axone. 2006;27(2):923.

  • 20

    Annual call for data: National Trauma Data Bank (NTDB). American College of Surgeons. Accessed August 20, 2020. https://www.facs.org/quality-programs/trauma/tqp/center-programs/ntd

    • Export Citation
  • 21

    Fantus RJ. Annual Report 2017: ICD-10. Bulletin. January 6, 2018. Accessed August 20, 2020. https://bulletin.facs.org/2018/01/annual-report-2017-icd-10/

    • Search Google Scholar
    • Export Citation
  • 22

    Harrell F. Regression modeling strategies. R package version 4.4-2. 2016. Accessed August 20, 2020. https://cran.r-project.org/web/packages/rms/index.html

    • Search Google Scholar
    • Export Citation
  • 23

    Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis. Springer Science & Business Media; 2013.

    • Search Google Scholar
    • Export Citation
  • 24

    Rozzelle CJ, Wofford JL, Branch CL. Predictors of hospital mortality in older patients with subdural hematoma. J Am Geriatr Soc. 1995;43(3):240244.

    • Search Google Scholar
    • Export Citation
  • 25

    Scotti P, Séguin C, Lo BWY, Antithrombotic agents and traumatic brain injury in the elderly population: hemorrhage patterns and outcomes. J Neurosurg. 2020;133(2):486495.

    • Search Google Scholar
    • Export Citation
  • 26

    Puffer RC, Yue JK, Mesley M, Long-term outcome in traumatic brain injury patients with midline shift: a secondary analysis of the Phase 3 COBRIT clinical trial. J Neurosurg. 2018;131(2):596603.

    • Search Google Scholar
    • Export Citation
  • 27

    Thompson HJ, Rivara FP, Jurkovich GJ, Evaluation of the effect of intensity of care on mortality after traumatic brain injury. Crit Care Med. 2008;36(1):282290.

    • Search Google Scholar
    • Export Citation
  • 28

    Schupper AJ, Berndtson AE, Smith A, Respect your elders: effects of ageing on intracranial pressure monitor use in traumatic brain injury. Trauma Surg Acute Care Open. 2019;4(1):e000306.

    • Search Google Scholar
    • Export Citation
  • 29

    Silveira MJ, Kim SYH, Langa KM. Advance directives and outcomes of surrogate decision making before death. N Engl J Med. 2010;362(13):12111218.

    • Search Google Scholar
    • Export Citation
  • 30

    Gordy S, Klein E. Advance directives in the trauma intensive care unit: do they really matter? Int J Crit Illn Inj Sci. 2011;1(2):132137.

    • Search Google Scholar
    • Export Citation
  • 31

    Rabinstein AA, Chung SY, Rudzinski LA, Lanzino G. Seizures after evacuation of subdural hematomas: incidence, risk factors, and functional impact. J Neurosurg. 2010;112(2):455460.

    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 383 383 383
PDF Downloads 186 186 186
EPUB Downloads 0 0 0