Targeted versus Empiric Embolization for Delayed Postpancreatectomy Hemorrhage: A Retrospective Study of 312 Patients.

PURPOSE: To assess the safety and clinical effectiveness of empiric embolization (EE) compared with targeted embolization (TE) in the treatment of delayed postpancreatectomy hemorrhage (PPH). MATERIALS AND METHODS: The data of patients with delayed PPH between January 2012 and August 2022 were analyzed retrospectively. In total, 312 consecutive patients (59.6 years ± 10.8; 239 men) were included. The group was stratified into 3 cohorts according to angiographic results and treatment strategies: TE group, EE group, and no embolization (NE) group. The χ2 or Fisher exact test was implemented for comparing the clinical success and 30-day mortality. The variables related to clinical failure and 30-day mortality were identified by univariable and multivariable analyses. RESULTS: Clinical success of transcatheter arterial embolization was achieved in 70.0% (170/243) of patients who underwent embolization. There was no statistical difference in clinical success and 30-day mortality between the EE and TE groups. Multivariate analyses demonstrated that malignant disease (odds ratio [OR] = 5.76), Grade C pancreatic fistula (OR = 7.59), intra-abdominal infection (OR = 2.54), and concurrent extraluminal and intraluminal hemorrhage (OR = 2.52) were risk factors for clinical failure. Moreover, 33 patients (13.6%) died within 30 days after embolization. Advanced age (OR = 2.59) and intra-abdominal infection (OR = 5.55) were identified as risk factors for 30-day mortality. CONCLUSIONS: EE is safe and as effective as TE in preventing rebleeding and mortality in patients with angiographically negative delayed PPH.

High intratumoral susceptibility signal grade on susceptibility-weighted imaging: a risk factor for hemorrhage after stereotactic biopsy.

OBJECTIVE: This study aimed to examine the association of preoperative intratumoral susceptibility signal (ITSS) grade with hemorrhage after stereotactic biopsy (STB). METHODS: The authors retrospectively reviewed 66 patients who underwent STB in their institution. Preoperative factors including age, sex, platelet count, prothrombin time-international normalized ratio, activated thromboplastin time, antiplatelet agent use, history of diabetes mellitus and hypertension, target location, anesthesia type, and ITSS data were recorded. ITSS was defined as a dot-like or fine linear low signal within a tumor on susceptibility-weighted imaging (SWI) and was graded using a 3-point scale: grade 1, no ITSS within the lesion; grade 2, 1-10 ITSSs; and grade 3, ≥ 11 ITSSs. Postoperative final tumor pathology was also reviewed. The association between preoperative variables and the size of postoperative hemorrhage was examined. RESULTS: Thirty-four patients were men and 32 were women. The mean age was 66.6 years. The most common tumor location was the frontal lobe (27.3%, n = 18). The diagnostic yield of STB was 93.9%. The most common pathology was lymphoma (36.4%, n = 24). The ITSS was grade 1 in 37 patients (56.1%), grade 2 in 14 patients (21.2%), and grade 3 in 15 patients (22.7%). Interobserver agreement for ITSS was almost perfect (weighted kappa = 0.87; 95% CI 0.77-0.98). Age was significantly associated with ITSS (p = 0.0075). Postoperative hemorrhage occurred in 17 patients (25.8%). Maximum hemorrhage diameter (mean ± SD) was 1.78 ± 1.35 mm in grade 1 lesions, 2.98 ± 2.2 mm in grade 2 lesions, and 9.51 ± 2.11 mm in grade 3 lesions (p = 0.01). Hemorrhage > 10 mm in diameter occurred in 10 patients (15.2%), being symptomatic in 3 of them. Four of 6 patients with grade 3 ITSS glioblastomas (66.7%) had postoperative hemorrhages > 10 mm in diameter. After adjusting for age, ITSS grade was the only factor significantly associated with hemorrhage > 10 mm (p = 0.029). Compared with patients with grade 1 ITSS, the odds of postoperative hemorrhage > 10 mm in diameter were 2.57 times higher in patients with grade 2 ITSS (95% CI 0.31-21.1) and 9.73 times higher in patients with grade 3 ITSS (95% CI 1.57-60.5). CONCLUSIONS: ITSS grade on SWI is associated with size of postoperative hemorrhage after STB.

Renal pelvis mucosal artery hemorrhage after percutaneous nephrolithotomy: a rare case report and literature review.

BACKGROUND: Following a percutaneous nephrolithotomy (PCNL) procedure, the most common complications are considered to be intraoperative and postoperative bleeding. Many patients with postoperative bleeding can be treated conservatively, causing the perirenal hematoma to resolve spontaneously. The major causes of severe postoperative bleeding are pseudoaneurysms, arteriovenous fistula, and segmental arterial injury. Typically, the first choice of treatment to manage severe bleeding complications is selective angioembolization (SAE) because of the very high success rate associated with this procedure. CASE PRESENTATION: This clinical case involves a 56-year-old man who underwent dual-channel PCNL treatment after diagnosing a left kidney staghorn stone and urinary tract infection. The operation was successful, with no apparent signs of bleeding. Tests revealed continued decreasing hemoglobin levels following the procedure. After the conservative treatment failed, renal angiography was performed immediately, indicating renal pelvis mucosal artery hemorrhage. In the three hours post-surgery, the SAE still failed to prevent bleeding. Further discussions led to formulating a new surgical plan using a nephroscope to enter the initial channel where hemostasis began. The hemostasis origin was found precisely in the mucosal artery next to the channel during the operation and was successfully controlled. CONCLUSIONS: This case reveals there is poor communication and inadequate discussions about the potential failures of an SAE procedure. Swift clinical decision-making is imperative when dealing with high-level renal trauma to prevent delays in surgery that can threaten the safety of patients.

Routine Early Computed Tomography Scanner After Laparoscopic Sleeve Gastrectomy in High-Risk Severely Obese Patients Is Effective for Bleeding or Hematoma Diagnosis but not for Staple-Line Leak Detection: a Prospective Study.

PURPOSE: Laparoscopic sleeve gastrectomy (LSG) is the most frequently performed bariatric procedure worldwide. Postoperative staple-line leak and intraabdominal hemorrhage can increase associated morbidity and mortality. The value of routine early computed tomography (CT) scanner examination in the early diagnosis of complications in high-risk severely obese patients undergoing LSG is studied. METHODS: This was a prospective, non-randomized study including all patients undergoing LSG in our department from 2014 to 2020. Patients presenting at least one potential risk factor for postoperative gastric leak and bleeding (as defined by the current literature) were included. Primary endpoint was the efficacy of postoperative day (POD) 2 CT-scanner examination in diagnosing these complications. RESULTS: One thousand fifty-one high-risk patients were included. Median age was 44 years. Early postoperative surgical complications occurred in 48 patients (4.5%): 25 (2.3%) intraabdominal hemorrhage and 23 (2.2%) staple-line leak. Early CT-scanner detected intraabdominal bleeding or hematoma in 22/25 patients, with 95.6% sensitivity (Youden's index = 0.95), while specificity was 100%, positive predictive value (PPV) 100%, and negative predictive value (NPV) 99.9%. Sensitivity of early postoperative CT-scanner was 43.4% (10/23 patients; Youden's index = 0.43) for staple-line leak detection, with specificity of 100%, PPV 100%, and NPV 98.7%. CONCLUSION: POD 2 CT-scanner in high-risk severely obese patients undergoing LSG is an excellent tool for early diagnosis of intraabdominal hemorrhage, but sensitivity remains low for staple-line leak detection. Close postoperative clinical follow-up of these patients is essential and any suspicion of postoperative surgical complication should motivate the performance of a CT-scanner.

Incidence and Contemporary Management of Delayed Bleeding Following Pancreaticoduodenectomy.

BACKGROUND: Delayed bleeding after pancreaticoduodenectomy (PD) is a life-threatening complication. However, the optimal management remains unclear. We summarize our experience of the management of delayed bleeding after PD and define the outcomes associated with different types of management. METHODS: All patients who underwent a PD between January 1987 and June 2020 at Johns Hopkins University were retrospectively reviewed. Delayed bleeding was defined as bleeding on or after postoperative day 5 following PD. Incidence, outcomes, and trends were reported. RESULTS: Among the 6201 patients that underwent PD, delayed bleeding occurred in 130 (2.1%) at a median of 12 days (IQR: 9, 24) postoperation. The pattern of bleeding was classified as intraluminal (51.5%), extraluminal (40.8%), and mixed (7.7%). A clinically relevant postoperative pancreatic fistula and an intraabdominal abscess preceded the delayed bleeding in 43.1% and 31.5% of cases, respectively. Arterial pseudoaneurysm or bleeding from peripancreatic vessels was the most common reason (54.6%) with the gastroduodenal artery being the most common source (18.5%). Endoscopy, angiography, and reoperation were performed as a first-line approach in 35.4%, 52.3%, and 6.2% of patients, respectively. The overall mortality was 16.2% and decreased over the study period (p < 0.01). CONCLUSIONS: Delayed bleeding following PD remains a life-threatening complication. The most common location of delayed bleeding is from the gastroduodenal artery. Angiography with embolization should be the initial approach for urgent bleeding with surgical re-exploration reserved for unstable patients or failed control of bleeding after interventional angiography or endoscopy.

Association of Satellite Sign with Postoperative Rebleeding in Patients Undergoing Stereotactic Minimally Invasive Surgery for Hypertensive Intracerebral Haemorrhage.

There are few studies regarding imaging markers for predicting postoperative rebleeding after stereotactic minimally invasive surgery (MIS) for hypertensive intracerebral haemorrhage (ICH), and little is known about the relationship between satellite sign on computed tomography (CT) scans and postoperative rebleeding after MIS. This study aimed to determine the value of the CT satellite sign in predicting postoperative rebleeding in patients with hypertensive ICH who undergo stereotactic MIS. We retrospectively examined and analysed 105 patients with hypertensive ICH who underwent standard stereotactic MIS for hematoma evacuation within 72 h following admission. Postoperative rebleeding occurred in 14 of 65 (21.5%) patients with the satellite sign on baseline CT, and in 5 of the 40 (12.5%) patients without the satellite sign. This difference was statistically significant. Positive and negative values of the satellite sign for predicting postoperative rebleeding were 21.5% and 87.5%, respectively. Multivariate logistic regression analysis verified that baseline ICH volume and intraventricular rupture were independent predictors of postoperative rebleeding. In conclusion, the satellite sign on baseline CT scans may not predict postoperative rebleeding following stereotactic MIS for hypertensive ICH.

Delayed Postoperative Hemorrhage Complicating Major Supramesocolic Surgery Management and Outcomes.

BACKGROUND: The place of surgery and interventional radiology in the management of delayed (> 24 h) hemorrhage (DHR) complicating supramesocolic surgery is still to define. The aim of the study was to evaluate outcomes of DHR using a combined multimodal strategy. METHODS: Between 2005 and 2019, 57 patients (median age 64 years) experienced 86 DHR episodes after pancreatic resection (n = 26), liver transplantation (n = 24) and other (n = 7). Hemodynamically stable patients underwent computed tomography evaluation followed by interventional radiology (IR) treatment (stenting and/or embolization) or surveillance. Hemodynamically unstable patients were offered upfront surgery. Failure to identify the leak was managed by either prophylactic stenting/embolization of the most likely bleeding source or surveillance. RESULTS: Mortality was 32% (n = 18). Bleeding recurrence occurred in 22 patients (39%) and was multiple in 7 (12%). Sentinel bleeding was recorded in 77 (81%) of episodes, and the bleeding source could not be identified in 26 (30%). Failure to control bleeding was recorded in 9 (28%) of 32 episodes managed by surgery and 4 (11%) of 41 episodes managed by IR (p = 0.14). Recurrence was similar after stenting and embolization (n = 4/18, 22% vs n = 8/26, 31%, p = 0.75) of the bleeding source. Recurrence was significantly lower after prophylactic IR management than surveillance of an unidentified bleeding source (n = 2/10, 20% vs. n = 11/16, 69%, p = 0.042). CONCLUSION: IR management should be favored for the treatment of DHR in hemodynamically stable patients. Prophylactic IR management of an unidentified leak decreases recurrence risks.

The death risk factors of patients undergoing re-exploration for bleeding or tamponade after isolated off-pump coronary artery bypass grafting: a case-control study.

BACKGROUND: The purpose of the study is to identify off-pump patients who are at higher risk of mortality after re-exploration for bleeding or tamponade. METHODS: We analyzed the data of 3256 consecutive patients undergoing isolated off-pump coronary artery bypass grafting (OPCABG) in our heart center from 2013 through 2020. Fifty-eight patients underwent re-exploration after OPCABG. The 58 patients were divided into death group and survival group according to their discharge status. Propensity score matching (PSM) was performed to analysis the risk factors of death. 15 pairs of cases of two groups were matched well. RESULTS: The mortality rate of patients underwent re-exploration after OPCABG for bleeding or tamponade was 27.59% (16/58). In the raw data, we found the patients in death group had higher body mass index (BMI) (P = 0.030), higher cardiac troponin T (cTnT) (P = 0.028) and higher incidence of heart failure before OPCABG (P = 0.003). After PSM, the levels of lactic acid before and after re-exploration (P = 0.028 and P < 0.001) were higher in death group. And the levels of creatinine (P = 0.002) and cTnT (P = 0.017) were higher in the death group after re-exploration. The death group had longer reoperation time (P = 0.010). In addition, the perioperative utilization rate of intra-aortic ballon pump (IABP) (P = 0.027), continuous renal replacement therapy (CRRT) (P < 0.001) and platelet transfusion (P = 0.017) were higher than survival group. CONCLUSIONS: The mortality rate of patients undergoing re-exploration for bleeding or tamponade after isolated OPCABG is high. More attention should be paid to patients with above risk factors and appropriate measures should be taken in time.

Hemorrhage Rates After Implantation and Explantation of Stereotactic Electroencephalography: Reevaluating Patients' Risk.

OBJECTIVE: Stereoelectroencephalography (sEEG), despite its established usefulness, has not been thoroughly evaluated for its adverse events profile. In this study, hemorrhage rates were evaluated both per patient and per lead placed not only in the immediate postoperative period, but also over the course of admission and after explantation when available. METHODS: This is a single-center retrospective study of pediatric and adult patients undergoing sEEG lead placement at a large urban hospital. All available postoperative imaging was reviewed for the presence of hemorrhage, including any imaging occurring throughout admission as well as within 1 month of lead explantation. Age and number of leads placed per procedure were compared using an unpaired t test assuming unequal variance. RESULTS: A total of 1855 leads were placed in 147 cases. The mean age was 30.4 ±15.0 and the male/female ratio was 47:53. 9 leads (0.49%) in 9 cases (6.12%) were involved with postimplantation hemorrhage occurring on postoperative day 0.44 on average. Postexplantation imaging was available for 45 cases. Seven leads (1.40%) in 7 cases (15.56%) were involved with postexplantation hemorrhage occurring on average on postoperative day 1.42. There was a significant difference in mean age between patients with postexplantation hemorrhage versus control (45.0 vs. 32.2; P = 0.0277). No cases of hemorrhage required surgical intervention and no patients had permanent neurologic deficit. CONCLUSIONS: Hemorrhage after sEEG lead implantation and explantation may be more common than previously reported. Consistent postexplantation imaging may be of clinical benefit in detecting hemorrhage that precludes patients from immediate discharge, particularly in older patients.

Hockey: a pain in the butt! Isolated superior gluteal artery rupture following blunt pelvic trauma - an unusual case.

Superior gluteal artery rupture is a rare complication of trauma but a significant one with potential for substantial morbidity and mortality. This case demonstrates the importance of early diagnosis and treatment of this injury pattern. Endovascular embolisation has become the most effective treatment for pelvic haemorrhage. Acknowledgement of this rare injury as a differential diagnosis is vital to facilitate rapid diagnosis and appropriate treatment.

Profunda femoris pseudoaneurysm rupture post dynamic hip screw fixation after physiotherapy.

This case discusses an elderly female who presented acutely with compromised profunda femoris pseudoaneurysm and massive haematoma five weeks after dynamic hip screw insertion for a left neck of femur fracture. The only precipitating factor leading to this presentation was ongoing physiotherapy. She was referred from a rehabilitation hospital to the nearest vascular surgical unit for acute and definitive surgical intervention. Post-operatively, she fared incredibly well, regaining her baseline level of functioning. History taking is complex in a patient with dementia. Clinical examination should follow with a focused approach to the site of recent operation and also where complications are likely to manifest when an alteration from baseline cognitive function is noted. This is of course in addition to the complete work up required from a holistic perspective with any acute deterioration. Imaging should be arranged and prompt referral made if a treatable acute cause is identified. It is imperative to involve family and/or next of kin if possible, but this should not impede prompt decision-making in the patient's best interests by the clinical team if delays are likely to occur.

Hepatic Artery Embolization for Postoperative Hemorrhage: Importance of Arterial Collateral Vessels and Portal Venous Impairment.

PURPOSE: To investigate the association between hepatic ischemic complications and hepatic artery (HA) collateral vessels and portal venous (PV) impairment after HA embolization for postoperative hemorrhage. MATERIALS AND METHODS: From October 2003 to November 2019, 42 patients underwent HA embolization for postoperative hemorrhage. HA collateral vessels were classified according to visualization after embolization (grade 1, none; grade 2, 1-4 segmental HA; and grade 3, ≥4 segmental HA). Transhepatic portal vein stent placements were performed in the same session for 5 patients (11.9%) with poor HA collateral vessels (grade 1 or 2) and compromised PV flow (>70% stenosis). Hepatic ischemic complications were analyzed for relevance to HA collateral vessels and PV compromise. RESULTS: After HA embolization, HA flow was found to be preserved (grade 3) through intra- and/or extrahepatic collateral vessels in 23 patients (54.8%), and hepatic complications did not occur regardless of PV flow status (0%). Of the 19 patients (45.2%) with poor HA collateral vessels (grade 1 or 2), segmental hepatic infarction occurred in 2 of 15 patients (13.3%) with preserved PV flow (10 naïve and 5 stented). The remaining 4 patients with poor HA collateral vessels and untreated compromised PV flow experienced multisegmental hepatic infarction (n = 3) or hepatic failure (n = 1) (100%) (P < .005). CONCLUSIONS: After HA embolization, preserved HA flow (≥4 segmental HA) lowered the risk of hepatic complications regardless of the PV flow. Based on these findings, transhepatic PV stent placement seems to be an effective intervention for the prevention of hepatic complications in cases of poor HA collateral vessels and compromised PV flow.

Early diagnosis and prediction of intracranial hemorrhage using dual-energy computed tomography after mechanical thrombectomy in patients with acute ischemic stroke.

PURPOSE: This study assesses the clinical value of dual-energy computed tomography (DECT) in the early diagnosis of intracranial hemorrhage and evaluates the risk of hemorrhagic transformation in patients with acute ischemic stroke (AIS) after mechanical thrombectomy. METHODS: Patients with AIS who have undergone thrombectomy with Solitaire stent and DECT within one hour after surgery were prospectively enrolled. Linear mixed energy images, virtual non-contrast (VNC) image, and iodine overlay map (IOM) were obtained. Routine CT scan was performed 24 h postoperatively. The sensitivity, specificity, positive and negative predictive values, and accuracy of DECT in the early diagnosis of intracranial hemorrhage was evaluated. The iodine concentration of intracranial lesions was measured by IOM with the follow-up results taken as reference. Receiver operating characteristic (ROC) analysis was performed to obtain the threshold of hemorrhagic transformation and increased bleeding. RESULTS: Among the 44 patients enrolled in this study, 25 (56.8 %) were diagnosed with simple extravasation of iodinated contrast agent, and 19 (43.2 %) showed intracranial hemorrhage in DECT. Compared with the follow-up CT 24 h after surgery, early diagnosis of postoperative intracranial hemorrhage using DECT demonstrated a sensitivity of 90.5 %, specificity of 100 %, positive predictive rate of 100 %, negative predictive rate of 92.0 %, and accuracy of 95.5 %. Among the 86 intracranial lesions that underwent iodine concentration measurement, 19 were diagnosed with hemorrhagic transformation or increased bleeding, and 67 were diagnosed without the aforementioned conditions. The sensitivity and specificity for differentiating the two groups were 73.7 % and 92.5 %, respectively, with a cut-off value of 2.7 mg/mL. CONCLUSION: DECT is clinically valuable in early diagnosis and prediction of intracranial hemorrhage after mechanical thrombectomy in AIS patients.

Imaging and surgical predictive factors for postoperative hemorrhage after partial nephrectomy and clinical results of trans-arterial embolization.

ABSTRACT: Partial nephrectomy (PN) has been established as the standard treatment for T1 renal tumors, and postoperative hemorrhage due to vascular complications is a rare but potentially life-threatening complication reported after PN. Thus, this study evaluated the imaging and surgical factors associated with postoperative hemorrhage after PN and the clinical results of trans-arterial embolization. A retrospective review of the institutional PN database was performed from May 2012 to January 2019, revealing that we performed 810 PN procedures at our institution. In total, 12 patients were referred to the interventional radiology department for vascular complications after the procedure. Patients with and without transarterial embolization (TAE) were age- and sex-matched with 56 patients. Preoperative imaging characteristics and operative details were considered. Univariable and multivariable analyses were used to test their eventual association with the occurrence of hemorrhage. Furthermore, renal functions at diagnosis, after operation or embolization for TAE cases, and at the last follow-up were recorded. A diagnosis of hemorrhage was made at a median of 4 (range, 0-25) days after surgery. The majority of patients (50%) presented with gross hematuria. T test revealed higher renal tumor-parenchyma contact area (TPA) (P = .0407), Length-A (P = .0136), Length-P (P = .0267), operation time (P = .0214) and estimated blood loss (P = .0043) in patients with hemorrhage than in controls. Binary logistic regression analysis identified TPA (P = .048) and estimated blood loss (P = .042) as independent predictors for postoperative hemorrhage with an area under the ROC curve of 0.705 (64% sensitivity and 79% specificity). In conclusion, the occurrence of hemorrhage after PN was associated with a larger TPA and more estimated blood loss during the procedure. In patients who underwent selective TAE, renal function remained comparable with that of controls.

Postpancreatectomy hemorrhage with negative angiographic findings: outcomes of empiric embolization compared to conservative management.

OBJECTIVE: To evaluate the efficacy of empiric embolization for postpancreatectomy hemorrhage (PPH) with negative angiographic signs of active bleeding. MATERIALS AND METHODS: A total of 100 patients (76 men, 24 women) who were diagnosed with PPH with angiographic findings revealing no signs of active bleeding from December 2013 to December 2019 were included in the study. The patients were divided into two groups according to whether the procedures were performed with or without empiric embolization in angiography (group of empiric embolization, N=47; group of no embolization, N=53). Data reflecting patients' characteristics, hemorrhagic details, classification of PPH grade, and postoperative complications were acquired. The rates of clinical success in hemostasis and mortality were compared between the group of empiric embolization and the group of no embolization. RESULTS: In the group of empiric embolization, the rate of clinical success in hemostasis and mortality were 61.7% and 27.7%, respectively. In the group of no embolization, the rates of clinical success in hemostasis and mortality were 39.6% and 13.2%, respectively. The rate of clinical success in hemostasis in the group of empiric embolization was significantly higher than that in the group of no embolization (p = 0.028). There was no statistically significant difference in mortality between the different groups (p = 0.071). CONCLUSION: The clinical success rate of hemostasis in patients with empiric embolization is higher than that in patients with no embolization. Empiric embolization may be an efficacious hemostatic treatment for PPH with angiographic findings revealing no signs of active bleeding.

Acute and subacute prostate MRI findings after MRI-guided transurethral ultrasound ablation of prostate cancer.

BACKGROUND: Magnetic resonance imaging (MRI)-guided transurethral ultrasound ablation (TULSA) is an emerging method for treatment of localized prostate cancer (PCa). TULSA-related subacute MRI findings have not been previously characterized. PURPOSE: To evaluate acute and subacute MRI findings after TULSA treatment in a treat-and-resect setting. MATERIAL AND METHODS: Six men with newly diagnosed MRI-visible and biopsy-concordant clinically significant PCa were enrolled and completed the study. Eight lesions classified as PI-RADS 3-5 were focally ablated using TULSA. One- and three-week follow-up MRI scans were performed between TULSA and robot-assisted laparoscopic prostatectomy. RESULTS: TULSA-related hemorrhage was detected as a subtle T1 hyperintensity and more apparent T2 hypointensity in the MRI. Both prostate volume and non-perfused volume (NPV) markedly increased after TULSA at one week and three weeks after treatment, respectively. Lesion apparent diffusion coefficient values increased one week after treatment and decreased nearing the baseline values at the three-week MRI follow-up. CONCLUSION: The optimal timing of MRI follow-up seems to be at the earliest at three weeks after treatment, when the post-procedural edema has decreased and the NPV has matured. Diffusion-weighted imaging has little or no added diagnostic value in the subacute setting.