PROPELLER diffusion weighted imaging for diagnosis of cholesteatoma in comparison with surgical and histopathological results: emphasis on false positivity and false negativity.

PURPOSE: To investigate the role of non-echo planar diffusion weighted imaging (DWI) using "periodically rotated overlapping parallel lines with enhanced reconstruction" (PROPELLER) sequence for the diagnosis of cholesteatoma compared to surgical and histopathological results in an attempt to determine the factors causing false negative and false positive diagnoses. METHODS: Patients who had PROPELLER DWI before ear surgery were retrospectively reviewed. The presence of a lesion with diffusion restriction on PROPELLER DWI was accepted as positive for cholesteatoma, and the results were compared to the intraoperative and histopathological findings. RESULTS: A total of 112 ears in 109 patients were reviewed. On PROPELLER DWI, a lesion with diffusion restriction was found in 101 (90.2%) ears, while in 11 (9.8%) of the patients, no diffusion restriction was found. Surgery and histopathological analysis revealed a cholesteatoma in 100 (89.3%) ears, while in 12 (10.7%) ears, no cholesteatoma was found surgically. There were 96 (85.7%) true positives, 7 (6.2%) true negatives, 5 (4.5%) false positives and 4 (3.6%) false negatives. The accuracy, sensitivity, specificity, positive predictive and negative predictive values of non-echo planar DWI were calculated to be 91.96%, 96%, 58.33%, 95.05%, and 63.64%, respectively. CONCLUSION: Non-echo planar DWI using PROPELLER sequence has high accuracy, sensitivity and positive predictive value and can be used for the detection of cholesteatoma. The external auditory canal, postoperative ears and small lesions should be evaluated with caution to avoid false results.

Diffusion-weighted imaging versus non-contrast magnetic resonance imaging in the diagnosis of acute appendicitis during pregnancy.

OBJECTIVE: The aim of this study was to evaluate the diagnostic performance of diffusion-weighted imaging compared to non-contrast magnetic resonance imaging in the differential diagnosis of acute appendicitis in pregnant patients. METHODS: A total of 72 pregnant patients with the suspicion of acute appendicitis who underwent magnetic resonance imaging combined with diffusion-weighted imaging examinations were enrolled in this retrospective study. Magnetic resonance imaging images (non-contrast and diffusion-weighted imaging sequences) were evaluated. Moreover, apparent diffusion coefficient ratios were estimated. The diagnostic performances of magnetic resonance imaging and diffusion-weighted imaging findings were statistically analyzed on the basis of surgical and follow-up results. RESULTS: Of 72 pregnant patients, 10 (14%) had acute appendicitis on magnetic resonance imaging and diffusion-weighted imaging. Among 10 patients with acute appendicitis, three (3/10) had perforation. diffusion-weighted imaging findings had higher sensitivity (90 versus 60%), negative predictive value (98.41 versus 93.94%), and accuracy (98.61 versus 94.44%) ratios compared to non-contrast magnetic resonance imaging in the diagnosis of acute appendicitis. There was one false-negative result on diffusion-weighted imaging. Diffusion restriction facilitated the detection of appendicitis. The apparent diffusion coefficient ratios were lower in acute appendicitis than in the normal appendix (0.70±0.19 versus 0.96±0.16) (p<0.05). CONCLUSION: With a shorter scan time and higher diagnostic accuracy, diffusion-weighted imaging can be useful for the early diagnosis of acute appendicitis and for planning appropriate management.

Diffusion-weighted imaging versus non-contrast magnetic resonance imaging in the diagnosis of acute appendicitis during pregnancy

SUMMARY OBJECTIVE: The aim of this study was to evaluate the diagnostic performance of diffusion-weighted imaging compared to non-contrast magnetic resonance imaging in the differential diagnosis of acute appendicitis in pregnant patients. METHODS: A total of 72 pregnant patients with the suspicion of acute appendicitis who underwent magnetic resonance imaging combined with diffusion-weighted imaging examinations were enrolled in this retrospective study. Magnetic resonance imaging images (non-contrast and diffusion-weighted imaging sequences) were evaluated. Moreover, apparent diffusion coefficient ratios were estimated. The diagnostic performances of magnetic resonance imaging and diffusion-weighted imaging findings were statistically analyzed on the basis of surgical and follow-up results. RESULTS: Of 72 pregnant patients, 10 (14%) had acute appendicitis on magnetic resonance imaging and diffusion-weighted imaging. Among 10 patients with acute appendicitis, three (3/10) had perforation. diffusion-weighted imaging findings had higher sensitivity (90 versus 60%), negative predictive value (98.41 versus 93.94%), and accuracy (98.61 versus 94.44%) ratios compared to non-contrast magnetic resonance imaging in the diagnosis of acute appendicitis. There was one false-negative result on diffusion-weighted imaging. Diffusion restriction facilitated the detection of appendicitis. The apparent diffusion coefficient ratios were lower in acute appendicitis than in the normal appendix (0.70±0.19 versus 0.96±0.16) (p<0.05). CONCLUSION: With a shorter scan time and higher diagnostic accuracy, diffusion-weighted imaging can be useful for the early diagnosis of acute appendicitis and for planning appropriate management.

Portal vein thrombosis risk factors in liver transplant candidates.

Background and Aim: Portal vein thrombosis (PVT) is particularly detected in advanced liver cirrhosis patients. We aimed to analyze the risk factors for PVT in liver transplant candidates. Materials and Methods: Dataset for consecutive 165 cirrhotic patients who were evaluated for liver transplantation (LT) were retrospectively analyzed. We sorted patients into two groups: patients with PVT and patients without PVT. Included variables were age, sex, etiology of liver disease, body mass index, MELD-Na score, Child-Pugh score, clinical variables reflecting portal hypertension, and hepatocellular carcinoma. Univariate and multivariate logistic regression analyses were used to identify risk factors of PVT. Results: Of 165 LT candidates, 46 had PVT (27.9%). Ascites, thrombocytopenia, history of variceal bleeding, and band ligation were risk factors for PVT in univariate analysis. In multivariate analysis, only a history of variceal bleeding (OR 3.45, 95% CI 1.02-11.6, p=0.046) significantly increased the risk of PVT. Conclusion: The previous history of variceal bleeding predicts PVT development in cirrhosis, suggesting that the severity of portal hypertension is a major predictive factor for PVT in patients with cirrhosis. Future prospective studies are needed to risk stratifying cirrhosis patients prior to LT for future PVT development and to define the prophylactic role of anticoagulation in these patients.

Development of Diaphragmatic Hernia in Patients with Penetrating Left Thoracoabdominal Stab Wounds.

BACKGROUND: This study aimed to investigate the consequences of repairing versus not repairing diaphragmatic injury caused by penetrating left thoracoabdominal stab wounds. METHODS: Diagnostic laparoscopy was performed to evaluate the left diaphragm in patients with penetrating left thoracoabdominal stab wounds who did not have an indication for emergency laparotomy. Patients who did not consent to laparoscopy were discharged without undergoing surgery. Post-discharge radiological images of patients who underwent diaphragmatic repair and radiological images of patients who could not undergo laparoscopy, both during hospitalization and after discharge, were evaluated and compared. RESULTS: Diagnostic laparoscopy was performed on 109 patients. Diaphragmatic injuries were detected and repaired in 32 (29.36%) of these patients. Seventeen patients were lost to follow-up. After a mean follow-up of 57.67 months, none of the remaining 15 patients developed a diaphragmatic hernia. On the other hand, 43 patients refused to undergo diagnostic laparoscopy. Twenty of them were lost from follow-up. The diaphragmatic injury was detected in seven of the remaining 23 patients (30.44%) during initial computed tomography (CT) examinations. In this group, the mean follow-up time was 42.57 months, and delayed diaphragmatic hernia developed in one patient (14.30%). Patients who underwent diaphragmatic repair were compared to patients who did not undergo diagnostic laparoscopy but had diaphragmatic injuries detected on their CT. No statistical differences were detected. CONCLUSIONS: Diaphragmatic injuries caused by penetrating stab wounds can sometimes heal spontaneously. However, diagnostic laparoscopy is still relevant for revealing and repairing possible diaphragmatic injuries.

Plantar Venous Pump Activity, Generalized Joint Hypermobility, and Foot Mobility in Ballerinas: A Case-Control Study.

CONTEXT: The plantar venous pump (PVP) is a vascular network located between the muscles of the foot arch. It has been suggested foot mobility is critical for PVP activation. OBJECTIVE: This study investigates the effect of generalized joint hypermobility (GJH) on foot mobility and PVP activity in ballerinas and to determine how a short-term warm-up exercise impacts these parameters in the presence of GJH. DESIGN: Prospective study. METHODS: Twelve ballerinas (age = 16.73 [1.8] y) and 15 non-dancer-matched controls (age = 16.31 [2.1] y) were included. When performing weight-bearing activities, venous return parameters (diameter, velocity, and flow volume) from the posterior tibial vein were measured using color and spectral Doppler ultrasonography. Foot mobility was determined using navicular drop test. The participants performed a 10-minute cycling exercise to reveal its effect on reducing tissue stiffness. All measurements were performed synchronously and twice-before and after the cycling exercise. RESULTS: The results showed the navicular height obtained before and after exercise in the ballerinas was significantly different compared with that obtained in the sitting position, and the values obtained in the 3 measurement positions in the nondancer group were significantly different from each other (P < .05). The difference between the venous diameter values measured before and after exercise and the reference value was significant in both groups (P < .001). There was a significant decrease in venous velocity following exercise in the nondancer group (P = .044). The venous volume values obtained after exercise were significantly different between the groups (P = .039). CONCLUSIONS: The GJH has no distinctive effect on foot mobility and PVP activity in ballerinas. The presence of GJH did not influence the effect of short-term warm-up exercise on foot mobility and PVP activity. We believe that measuring the navicular drop and venous return with a more precise synchronization may provide additional information regarding the link between arch tension and venous return.

Outcomes in Sonographically Categorized Complex Cystic Breast Lesions.

OBJECTIVES: To investigate outcomes of patients with complex cystic breast lesions and to determine malignancy rates for each sonographic category. MATERIAL AND METHODS: A total of 124 women with a mean age of 45 were enrolled in this retrospective study. Complex cysts were classified into four categories according to sonographic findings (thick septa/wall, intramural nodule, solid/cystic ratio) by two radiologists. Size, margin, short-to-long axis ratio, and vascularity were also noted. The relationship between sonographic features and histopathological/follow-up results was evaluated. RESULTS: Of 124 complex cysts, 11 were malignant and 113 were benign. There were 31 with thick septa/wall (category 1), 14 with intramural nodule (category 2), 60 with solid/cystic ratio < 1 (category 3), and 19 with solid/cystic ratio > 1 (category 4). Among 11 malignant tumors, six intraductal papillary carcinomas and three invasive ductal carcinomas were in category 3, whereas one invasive ductal carcinoma was in category 1 and one was in category 4. Most of malignant lesions (9/11, 82%) were in category 3. Most of benign lesions (96/113, 85%) were fibroadenomatoid/fibrocystic changes. The presence of vascularity (96%), irregular margin (94%), and short-to-long axis ratio ≥ 1 (94%) had higher accuracy rates for diagnosis of malignancy. CONCLUSION: The malignancy rate was 8.8% in our study. Besides internal morphological features, other sonographic features should be considered in the evaluation of complex cystic breast lesions. Due to substantial risk of malignancy and the overlapping sonographic features of benign and malignant lesions, tissue sampling of all complex cysts should be performed.

The Role of CO-RADS Scoring System in the Diagnosis of COVID-19 Infection and its Correlation with Clinical Signs.

BACKGROUND: Computed tomography (CT) evaluation systematics has become necessary to eliminate the difference of opinion among radiologists in evaluating COVID-19 CT findings. INTRODUCTION: The objectives of this study were to evaluate the efficiency of CO-RADS scoring system in our patients with COVID-19 as well as to examine its correlation with clinical and laboratory findings. METHODS: The CO-RADS category of all patients included in the study was determined by a radiologist who did not know the rtRT-PCR test result of the patients, according to the Covid-19 reporting and data system of Mathias Prokop et al. Results: A total of 1338 patients were included. CT findings were positive in 66.3%, with a mean CO-RADS score of 3,4 ± 1,7. 444 (33.1%) of the patients were in the CO-RADS 1-2, 894 (66.9%) were in the CO-RADS 3-5 group. There were positive correlations between CO-RADS score and age, CMI, hypertension, diabetes mellitus, chronic pulmonary diseases presence of symptoms, symptom duration, presence of cough, shortness of breath, malaise, CRP, and LDH, while CORADS score was negatively correlated with lymphocyte count. The results of the ROC analysis suggested that those with age ≥40 years, symptom duration >2 days, CMI score >1 and/or comorbid conditions were more likely to have a CO-RADS score of 3-5. CONCLUSION: The CO-RADS classification system is a CT findings assessment system that can be used to diagnose COVID-19 in patients with symptoms of cough, shortness of breath, myalgia and fatigue for more than two days.

Chest computed tomography findings of 1271 patients with COVID-19 pneumonia and classifications with different age groups: a descriptive study from Istanbul, Turkey.

OBJECTIVE: The objectives of this study were to describe lung computed tomography findings of patients with COVID-19 diagnosed by real-time reverse transcription polymerase chain reaction test, investigate whether the findings differ regarding age and gender, and evaluate the diagnostic performance of chest computed tomography based on the duration of symptoms at the time of presentation to the hospital. METHODS: From March 11 to May 11, 2020, 1271 consecutive patients (733 males and 538 females) were included in this retrospective, cross-sectional study. Based on age, patients were divided into five separate subgroups. Then based on the duration of symptoms, patients were divided into five separate phases. The presence of lung lesion(s) and their characteristics, distribution patterns, and the presence of concomitant pleural thickening/effusion and other findings (malignancy, metastasis, chronic obstructive pulmonary disease, interstitial lung disease, bronchiectasis, bronchiectasis, cardiomegaly, pericardial effusion) were evaluated by five radiologists independently. RESULTS: The "normal lung computed tomography finding" was the most common chest CT finding (37%), followed by ground-glass opacity (31%). Regardless of the shape of the lesion, the distribution features were significant (peripheral, subpleural, and lower lobe distribution) (p<0.05). The presence of pleural thickening posteriorly and adjacent to the lesion was statistically different in groups 1-3 (p<0.05). Other concomitant pathologies, except pulmonary congestion, did not suppress the typical findings of COVID-19. CONCLUSION: Chest computed tomography findings were mostly normal in the early phase (P1). Therefore, it may be appropriate to perform the first computed tomography screening of COVID-19 after 6 days to decrease the radiation exposure.

Chest computed tomography findings of 1271 patients with COVID-19 pneumonia and classifications with different age groups: a descriptive study from Istanbul, Turkey

SUMMARY OBJECTIVE: The objectives of this study were to describe lung computed tomography findings of patients with COVID-19 diagnosed by real-time reverse transcription polymerase chain reaction test, investigate whether the findings differ regarding age and gender, and evaluate the diagnostic performance of chest computed tomography based on the duration of symptoms at the time of presentation to the hospital. METHODS: From March 11 to May 11, 2020, 1271 consecutive patients (733 males and 538 females) were included in this retrospective, cross-sectional study. Based on age, patients were divided into five separate subgroups. Then based on the duration of symptoms, patients were divided into five separate phases. The presence of lung lesion(s) and their characteristics, distribution patterns, and the presence of concomitant pleural thickening/effusion and other findings (malignancy, metastasis, chronic obstructive pulmonary disease, interstitial lung disease, bronchiectasis, bronchiectasis, cardiomegaly, pericardial effusion) were evaluated by five radiologists independently. RESULTS: The "normal lung computed tomography finding" was the most common chest CT finding (37%), followed by ground-glass opacity (31%). Regardless of the shape of the lesion, the distribution features were significant (peripheral, subpleural, and lower lobe distribution) (p<0.05). The presence of pleural thickening posteriorly and adjacent to the lesion was statistically different in groups 1-3 (p<0.05). Other concomitant pathologies, except pulmonary congestion, did not suppress the typical findings of COVID-19. CONCLUSION: Chest computed tomography findings were mostly normal in the early phase (P1). Therefore, it may be appropriate to perform the first computed tomography screening of COVID-19 after 6 days to decrease the radiation exposure.

Synchronous pancreas and liver resection during CRS ± HIPEC. Results of 35 consecutive patients.

AIM: To evaluate the results of synchronous pancreas and liver resection in patients with metastatic carcinoma accompanied by peritoneal carcinomatosis (PC) who underwent cytoreductive surgery (CRS)/hyperthermic intraperitoneal chemotherapy (HIPEC). MATERIAL AND METHODS: Overall, 35 patients with PC who underwent CRS/HIPEC at Umraniye Education and Research Hospital owing to different etiological causes were evaluated for operation procedure, age, sex, comorbid diseases, ASA score, systemic chemo/radiotherapy, duration of surgery, chemotherapy (CT) protocol applied in HIPEC, amount of bleeding, synchronous hepatobiliary resection interventions during operation and short/long-term mortality and morbidity. RESULTS: Fifteen patients were male and 20 were female. Mean age was 58.375 ± 13.02 (range, 27-85) years. CRS/HIPEC was performed in 12 patients (34.3%), whereas only CRS was performed in 23 (65.7%) patients. The lowest and highest perioperative bleeding amount was 50 cc and 2000 cc, respectively, and 6 (17.1%) patients had intraoperative bleeding of ≥1000 cc. Synchronous isolated pancreatic resection, hepatic + pancreatic resection and isolated liver resection was performed in 3 (8.6%), 5 (14.3%) and 27 (77.1%) patients treated with CRS/HIPEC. Two patients died within postoperative day 30, and the mortality rate was 5.7%. CONCLUSION: Hepatopancreatobiliary surgery is a difficult intervention and is associated with serious morbidities. Synchronisation of hepatopancreatobiliary surgery with CRS/HIPEC may be challenging for the surgeon and clinician. Here we emphasise that this combination can be performed safely by an experienced team with a multidisciplinary approach and good patient preparation. KEY WORDS: Cytoreductive surgery, HIPEC, Hepatopancreatobiliary surgery, Peritoneal carcinomatosis.

Role of Interface Sign and Diffusion-Weighted Magnetic Resonance Imaging in Differential Diagnosis of Exophytic Renal Masses.

PURPOSE: We aimed to investigate the role of interfaces of exophytic solid and cystic renal masses on magnetic resonance imaging (MRI) and the added value of diffusion-weighted imaging in differentiating benign from malignant lesions. METHODS: The Institutional Review Board approved this retrospective study, and informed consent was waived. A total of 265 patients (109 [41%] women and 156 [59%] men) with a mean age of 57 ± 12 (standard deviation) years were enrolled in this study. Preoperative MRI (n = 238) examinations of patients with solid or cystic renal masses and MRI (n = 27) examinations of patients with Bosniak IIF cysts without progression were reviewed. Solid/cystic pattern, interface types and apparent diffusion coefficient (ADC) values were recorded by 2 radiologists. The diagnostic performance of combining normalized ADC values with interface sign were evaluated. RESULTS: Among 265 renal lesions (109 cystic and 156 solid), all malignant lesions (n = 192) had a round interface. No malignant lesions showed an angular interface. For prediction of benignity in cystic lesions, sensitivity (82.86% vs 56.16%), negative predictive value (92.50% vs 85.71%), and accuracy (94.50% vs 87.92%) ratios of angular interface were higher compared to all (solid plus cystic) lesions. The best normalized ADC cutoff values for predicting malignancy in lesions with round interface were as follows: for all (solid plus cystic), ≤ 0.75 (AUROC = 0.804); solid, ≤ 0.6 (AUROC = 0.819); and cystic, ≤ 0.8 (AUROC = 0.936). CONCLUSIONS: Angular interface can be a predictor of benignity for especially cystic renal masses. The evaluation of interface type with normalized ADC value can be an important clue in differential diagnosis especially in patients avoiding contrast.

Misleading findings of liver-specific MR contrast agent for radiological diagnosis of cysto-biliary communication in hydatid cysts.

OBJECTIVE: To investigate the effectiveness of gadoxetic acid-enhanced magnetic resonance cholangiopancreatography (CE-MRCP) and T2-weighted half-Fourier acquisition single-shot turbo spin-echo (T2W HASTE) sequences for diagnosis of cysto-biliary communication in hydatid cysts compared to surgical results. METHODS: Preoperative abdominal magnetic resonance imaging examinations of patients who underwent surgery for hepatic hydatid cysts were reviewed by two radiologists retrospectively. A total of 45 patients with hydatid cysts were included. Of 45, 27 also had CE-MRCPs. T2W HASTE sequences and CE-MRCPs were investigated separately for cysto-biliary communication. The relationship between radiological and surgical results was analyzed. The interobserver agreement was evaluated. RESULTS: Of 45 hydatid cysts, there were surgically proven 21 cysts without biliary communications and 24 cysts with biliary communications. All cysts with biliary communications were shown on T2W HASTE sequences. There was no leakage of gadoxetic acid into these cysts (n = 24). Sensitivity, specificity, negative predictive value, positive predictive value (PPV), and accuracy of diagnosis of cyst with biliary communication on T2W HASTE sequences was 100%, 63.64%, 100%, 66.67%, and 78.95%, respectively. Specificity (77.78%), PPV (87.50%), and accuracy (91.30%) were increased in ≥ 10 cm cysts. There was almost perfect interobserver agreement (K = 0.81-1.00). CONCLUSION: Leakage of gadoxetic acid inside the cyst indicates biliary communication. However, the lack of leakage does not rule out cysto-biliary communication. When biliary communication is clearly shown on T2W HASTE sequences, it should be reported as cysto-biliary communication even if there is no leakage of gadoxetic acid into the cyst on CE-MRCP.

Sinistral Portal Hypertension Due to Pancreatic Hydatid Cyst.

Hydatid disease is caused by Echinococcus granulosus. Hydatid cysts are commonly located in the liver and lungs. The occurrence of pancreatic hydatid cysts is very rare, even in endemic areas. Sinistral portal hypertension, which is rarely seen, occurs when a pathological process causes splenic vein occlusion. A 26-year-old male patient presented with abdominal pain. He had a history of operation for hydatid cyst of the lung 15 years ago. A left thoracotomy incision scar was observed during his physical examination. Laboratory findings revealed no abnormalities. Abdominal ultrasonography revealed a 96×69-mm lobular, contoured, well-circumscribed cystic lesion with thickened septation. Abdominal magnetic resonance imaging revealed a 100×76-mm smooth, bordered cystic lesion containing septations in the body and tail of the pancreas compressing the splenic artery and vein, causing sinistral portal hypertension. Dilatation was noted in the left gastroepiploic vein. The patient underwent cystotomy. Pancreatic fistula developed during the postoperative follow-up. The patient was discharged in 20 days without postoperative complications. No complications were observed during the follow-up period of 7 months. Surgery should be considered as a more conservative approach.

Evaluation of diaphragm in penetrating left thoracoabdominal stab injuries: The role of multislice computed tomography.

INTRODUCTION: Penetrating left thoracoabdominal stab injuries are accompanied by diaphragmatic injury in 25-30% of cases, about 30% of which later develop into diaphragmatic hernia. This study aimed to determine the role of multislice computed tomography in the evaluation of left diaphragm in patients with penetrating left thoracoabdominal stab wounds. MATERIALS AND METHODS: This study reviewed penetrating left thoracoabdominal stab injuries managed in our clinic between April 2009 and September 2014. The thoracoabdominal region was defined as the region between the sternum, fourth intercostal space, and arcus costa anteriorly and the vertebra, lower tip of scapula, and the curve of the last rib posteriorly. Unstable cases and cases with signs of peritonitis were operated with laparotomy; the remaining patients were closely monitored. Forty-eight hours later, a diagnostic laparoscopy was applied to evaluate the left hemidiaphragma in asymptomatic patients who did not need laparotomy. The preoperatively obtained multislice thoracoabdominal computed tomography images were retrospectively examined for the presence of left diaphragm injury. Then, operative and tomographic findings were compared. RESULTS: This study included a total of 43 patients, 39 (91%) males and 4 (9%) females of mean age 30 years (range 15-61 years). Thirty patients had normal tomography results, whereas 13 had left diaphragmatic injuries. An injury to the left diaphragm was detected during the operation in 9 (1 in laparotomy and 8 in diagnostic laparoscopy) of 13 patients with positive tomography for left diaphragmatic injury and 2 (in diagnostic laparoscopy) of 30 patients with negative tomography. Multislice tomography had a sensitivity of 82% (95% CI: 48-98%), a specificity of 88% (71-96%), a positive predictive value of 69% (39-91%), and a negative predictive value of 93% (78-99%) for detection of diaphragmatic injury in penetrating left thoracoabdominal stab injury. CONCLUSIONS: Although diagnostic laparoscopy is the gold standard for diaphragmatic examination in patients with penetrating left thoracoabdominal stab wounds, multislice computed tomography is also valuable for detecting diaphragmatic injury.