- Email: [email protected]
Traumatic rupture of the diaphragm: experience with 65 patients
Injury, Int. J. Care Injured 34 (2003) 169–172
Traumatic rupture of the diaphragm: experience with 65 patients Peter Mihos a , Konstantinos Potaris a,∗ , John Gakidis a , John Paraskevopoulos b , Panagiotis Varvatsoulis b , Basil Gougoutas c , George Papadakis b , Eleftherios Lapidakis c a Department of Thoracic Surgery, General Hospital of Attica ‘KAT’, 2 Nikis Street, Kifissia 14561, Greece First Department of General Surgery, General Hospital of Attica ‘KAT’, 2 Nikis Street, Kifissia 14561, Greece Second Department of General Surgery, General Hospital of Attica ‘KAT’, 2 Nikis Street, Kifissia 14561, Greece
b c
Accepted 18 October 2002
Abstract Traumatic diaphragmatic rupture is reported with increasing frequency and is associated with high morbidity and mortality. The purpose of this study was to present our experience with the management of this injury. Sixty-five patients with TDR were treated in our hospital between January 1989 and May 2000. They were 54 men (83%) and 11 women (17%). Mean age was 36.57 years (range 15–76 years). Rupture of the diaphragm was left-sided in 43 patients (66%), right-sided in 21 (32%), and bilateral in 1 (1.5%). Blunt trauma accounted for the injuries of 52 patients (80%). Early diagnosis was obtained in 57 patients (88%). The diagnosis was established preoperatively in 17 patients (26%), and intra-operatively in 48 (74%). Multiple associated injuries were observed in 62 patients (95%). Postoperative complications were observed in seven patients (11%), and the overall mortality was 14%. Injury severity score (ISS) and haemorrhagic shock upon admission strongly influence the outcome. A high index of suspicion and a thorough examination of both hemi diaphragms during laparotomy is recommended in order to avoid early or late complications. © 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction Sennertus was the first to describe a traumatic diaphragmatic hernia in 1541. Ambroise Pare diagnosed the first diaphragmatic rupture during autopsy, in 1579 [1,2]. Bowditch made an ante mortem diagnosis of a traumatic diaphragmatic rupture (TDR) in 1853 [3]. The first collective review of traumatic diaphragmatic rupture was published in 1951 and numerous studies have discussed traumatic diaphragmatic rupture since that time [4]. Acute diaphragmatic rupture occurs in 1–7% of major blunt trauma victims and in 10–15% of patients with penetrating trauma to the lower chest [1]. In conservatively managed patients the rate of initially missed diaphragmatic injuries ranges from 12 to 66%, and they may even be overlooked at laparotomy [1,5,6]. Diagnosis of a diaphragmatic injury requires a high index of suspicion. Delayed diagnosis of TDR is associated with increased morbidity and mortality and laparotomy has been suggested for all penetrating injuries of the lower thorax and upper abdomen [7]. ∗ Corresponding author. Tel.: +30-2-10-601-2587; fax: +30-2-10-231-3899. E-mail address: [email protected] (K. Potaris).
The aim of this retrospective study was to review the experience of our hospital with the management of TDR in order to identify incidence, associated morbidity and mortality, predictors of outcome, and factors contributing to diagnostic delay.
2. Material and methods The ‘KAT’ District General Hospital has a trauma center considered to be the busiest in Greece. The hospital has approximately 12,000 admissions for trauma per year. Blunt and penetrating trauma to the torso accounts for approximately 3000 admissions per year. We reviewed the notes and radiographs of 65 patients with a diagnosis of TDR treated between January 1989 and May 2000. The incidence, side of TDR, mechanism of injury, associated morbidity and mortality, and predictors of outcome were retrospectively evaluated. The majority of these patients who sustained major associated injuries were motor vehicle occupants. The variables studied and correlated with outcome included patient age, haemodynamic status upon admission, injury severity score (ISS), and time to diagnosis.
0020-1383/03/$ – see front matter © 2003 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 0 2 ) 0 0 3 6 9 - 8
170
P. Mihos et al. / Injury, Int. J. Care Injured 34 (2003) 169–172
Statistical analysis: continuous variables were compared using the unpaired Student’s t-test and differences between categorical variables were assessed using the χ 2 -test. A P-value of <0.05 was considered statistically significant.
3. Results During the specified period, 65 patients were diagnosed with TDR. Estimated incidence was 0.2% of blunt and penetrating trauma to the torso. The study identified 54 men (83%) and 11 women (17%) ranging in age from 15 to 76 years (mean 36.6 years). Causes of injury included traffic accidents (n = 51, 78%), gunshot wounds (n = 7, 11%), stab wounds (n = 6, 9%), and a fall from height (n = 1, 1.5%). TDR was left-sided in 43 patients (66%), right-sided in 21 (32%), and bilateral in 1 (1.5%). The majority of road accident induced TDR was left-sided (36 out of 51 patients, 71%). Gunshot wounds injured the right hemi diaphragm in four out of seven patients (57%). Stab wounds injured the left hemi diaphragm in four out of six patients (66.67%). Detailed data on the cause of injury and the location of TDR are given in Table 1. Diagnosis of TDR was established in less than 24 h in 57 patients (88%). In the remaining eight patients diagnostic delay ranged from 1 to 5 days with the exception of one patient who developed intra-thoracic herniation of abdominal viscera 8 years later. In 48 patients, TDR was diagnosed intra-operatively (74%) and in 17 preoperatively (26%). Diagnostic methods included chest X-ray, computerized tomography (CT) scan of the chest and upper abdomen, and upper gastrointestinal (GI) contrast study. The drainage of gastric contents via a chest tube disclosed a left TDR and a presumed iatrogenic rupture of the fundus of the stomach in one of our patients. They were both repaired during a revision laparotomy. He had undergone decortication for empyema developed after splenectomy and suturing of liver lacerations caused by a road traffic accident. 3.1. Diagnostic methods Plain chest X-rays were diagnostic of TDR in only 2 out of 65 patients (3%) but indicative of TDR in 24 patients (37%). Table 1 Cause of injury and location of traumatic diaphragmatic rupture sustained in 65 patients No. of patients
RH
LH
Bilateral
TA Falls Gunshots Stab wounds
51 1 7 6
15 (29%) 0 4 (57.1%) 2 (33.3%)
36 (71%) 1 2 (28.6%) 4 (66.7%)
0 0 1 (14.28%) 0
Total
65
21 (32.3%)
43 (66.1%)
1 (1.5%)
TA: traffic accidents; RH: right hemi diaphragm; LH: left hemi diaphragm.
They underwent either computerized tomography scan of the chest and upper abdomen or gastrografin study of the upper gastrointestinal tract, following nasogastric tube insertion. The decision depended upon the experience of the on call surgeons. CT scan of the chest and upper abdomen was performed in 10 haemodynamically stable patients. It was diagnostic of TDR in eight of them and indicative of abdominal organ herniation into the chest in the remaining two patients. Upper GI contrast study following insertion of a nasogastric tube and injection of gastrografin was performed in 14 patients. It was diagnostic of TDR in seven patients and false negative in the remainder. Diagnostic peritoneal lavage (DPL) was performed in 55 patients suspected of intra-abdominal haemorrhage and was positive for bleeding in 48 of them (87%). An intra-operative diagnosis of TDR was made in these 48 patients who underwent emergency laparotomy. DPL was indirectly diagnostic of TDR in one patient (1.5%). He had undergone chest tube insertion for haemothorax and the lavage fluid was drained through the chest tube. 3.2. Surgery The majority of our patients (n = 62, 95.4%) had multiple significant associated injuries, which are shown in Table 2. Except for the patients who had sustained both liver and spleen lacerations we observed that patients who had sustained right-sided TDR had liver injury and patients who had sustained left-sided TDR had spleen injury. Haemorrhagic shock occurred in 22 patients. Sixteen of these patients survived the operation while six died in the operating room because of profuse bleeding. Three others died later with adult respiratory distress syndrome (ARDS) in the intensive care unit (ICU). Blunt injury accounted for five of the six deaths. One patient died due to profuse bleeding, following Table 2 Multiple associated injuries observed in 62 patients sustained traumatic diaphragmatic rupture Type of injury
No. of patients
Head injury Lung laceration Pneumothorax Haemothorax Pericardial laceration Rib fractures Extremity fractures IVC laceration Spleen laceration Liver laceration Bowel laceration Kidney rupture Stomach rupture Gallbladder rupture Ovary avulsion
4 9 3 16 1 15 20 3 26 24 14 3 7 2 1
IVC: inferior vena cava.
P. Mihos et al. / Injury, Int. J. Care Injured 34 (2003) 169–172
171
Table 3 Risk factors influenced the outcome in 65 patients sustained traumatic diaphragmatic rupture
Age (years) Haemorrhagic shock (no. of patients) Early diagnosis (<24 h) (no. of patients) Late diagnosis (>24 h) (no. of patients) ISS
Survivors (n = 56)
Fatalities (n = 9)
P-value
Mean: 37.07, range: 15–68 16 (28.6%) 49 (87.5%) 7 (12.5%) Mean: 17.67, range: 9–32
Mean: 32.89, range: 29–72 6 (66.7%) 8 (88.9%) 1 (11.1%) Mean: 41.33, range: 29–72
0.330 (NS) 0.0247 (<0.05) 0.91 (NS) 0.90 (NS) 0.0004 (<0.001)
ISS: injury severity score; NS: not significant.
penetrating injury caused by a lead shot. He had TDR with associated liver and inferior vena cava lacerations. A similar but not lethal injury caused a bilateral diaphragmatic rupture to another patient who sustained major liver lacerations but no heart or great vessel injury. Two other patients underwent emergency thoracotomy because of haemothorax caused by laceration of the internal mammary artery in one, and the subclavian artery in the other. The mammary artery was ligated and the subclavian artery was primarily repaired with a polypropylene suture. They were both successfully resuscitated. Herniation of abdominal contents was observed in eight patients (12%) and included stomach in three, stomach and small bowel in one, stomach, colon and omentum in one, left colon flexure, small bowel and spleen in one, left colon flexure, left lobe of liver, spleen and upper stomach in one, and upper stomach, left colon flexure and spleen in one. The diaphragm was repaired via laparotomy in 39 patients (60%), thoracotomy in 13 (20%), and laparotomy with thoracotomy in 13 (20%). The choice depended on the clinical and radiological scenario and the associated injuries. Both interrupted and running techniques with non-absorbable suture were used. No prosthetic material was needed to close the defects. In all but one patient, surgery was performed for an acute diaphragmatic injury. Surgical treatment of the associated injuries was performed accordingly. 3.3. Outcome Postoperative complications were encountered among seven patients (11%). These included thoracic empyema and gastropleural fistula in one patient, thoracic empyema and paralytic ileus in one, mechanical ileus in one, paralytic ileus in one and pneumonia in three. They were all treated successfully either surgically or conservatively. Nine patients died (14%), six in the operating room due to non-reversible haemorrhagic shock, and the remaining three died from ARDS in the ICU. Mean hospital stay was 19.22 days (range 1–112 days). Twenty-three patients (37%) were treated in ICU. Their ICU stay ranged from 1 to 54 days with a mean of 7.3 days. Follow-up ranged from 1 to 45 months. Thirty-five of the survivors were lost to follow-up (62.5%) either because they were treated in different departments or because they failed follow-up. The remaining 21 patients had no further sequelae.
3.4. Risk factors In order to investigate risk factors influencing the outcome, we compared age, haemodynamic status, early (<24 h) or late diagnosis (>24 h), and ISS. We observed a statistically significant risk associated with haemorrhagic shock (P < 0.05), and a very significant statistically risk associated with ISS (P < 0.001). Neither age nor delay in diagnosis posed a statistically significant risk to the outcome of our patients. Observations and statistics are outlined in Table 3.
4. Comment Traumatic diaphragmatic rupture results from blunt or penetrating thoraco-abdominal trauma. It is common and regarded as a marker of trauma severity [1]. A large collective review, in 1995, suggested that 75% of the injuries to the diaphragm are caused by blunt trauma and 25% by penetrating trauma [2]. The incidence of TDR is reported to be between 0.8 and 7% when associated with blunt trauma, and between 10 and 15% when associated with penetrating trauma [8]. In a literature review in 1988, it was reported that in North American series blunt trauma accounts for 10–30% of traumatic diaphragmatic ruptures, whereas in Western Europe series blunt trauma accounts for 80–100% of TDR [9]. The true incidence of TDR is unknown because in 7–66% of major trauma victims, the diagnosis is missed. This is particularly true for ruptures of the right hemi diaphragm [1,2]. In our series, the incidence of TDR after blunt and penetrating truncal trauma was 0.2%. The principle cause of TDR was blunt trauma following road accident (77%). In blunt truncal trauma, most ruptures occur on the left posterolateral aspect of the diaphragm. This is a structurally weak area as it originates from the pleuroperitoneal membrane. The right diaphragm is congenitally stronger than the left and is partially protected by the liver, which can dissipate pressure over a large area. There is a preponderance of left-sided diaphragmatic ruptures following blunt trauma reported to be between 68.5 and 87% [2,4,10–13]. In our series, left-sided TDR occurred in 43 patients (66%). Blunt trauma accounted for the ruptures in 37 out of these 43 patients (86%).
172
P. Mihos et al. / Injury, Int. J. Care Injured 34 (2003) 169–172
There is no doubt in the literature as regards the need for aggressive operative treatment of TDR. This is due to early and late associated morbidity and mortality of this injury. We obtained an early diagnosis (less than 24 h) of TDR in 57 patients (88%). We noted 80% sensitivity of the CT scan of the chest and upper abdomen and 50% sensitivity for the upper GI contrast study. Plain chest X-ray was indicative of TDR in 24 patients (37%). Diagnostic methods that have been reported to be useful in the evaluation of TDR include plain chest X-ray, upper GI contrast study, DPL, fluoroscopic evaluation of diaphragmatic motion, ultrasound, CT scan, MRI, intra-peritoneal injection of radioisotopes, laparoscopy and video-assisted thoracic surgery. However, there are no studies to compare sensitivity, specificity and diagnostic accuracy of these modalities in the evaluation of TDR. The two modalities most commonly used, chest X-ray and CT scan are diagnostic in 30–50% of the cases. It has been reported that only 25–50% of the initial chest X-rays are diagnostic of TDR [1,5,7,8,14]. Nevertheless, in a study from Finland, chest X-rays on admission supported the diagnosis in 70% of cases and the authors concluded that a chest X-ray upon admission is the best diagnostic aid in the evaluation of TDR [15]. Sensitivity and specificity for CT scan ranges between 33–83 and 76–100%, respectively. Helical CT scanning with axial, sagittal and coronal reformations is reported to reach a sensitivity of 50% and 78% in diagnosing right- and left-sided diaphragmatic injuries, respectively [5,8,14,16]. CT scan is probably the diagnostic modality of choice in a resuscitated stable patient. DPL is reported to have a sensitivity of 66–75% and this probably relates to the degree of the associated injuries rather than diagnostic accuracy [1,7]. It was indirectly diagnostic for TDR in one of our patients (1.5%). In conclusion, an early diagnosis of TDR can easily be missed in the acute trauma setting despite the technological equipment and available diagnostic modalities. Chest X-rays or CT scans may be indicative. A thorough examination of both hemi diaphragms is recommended for all patients undergoing emergency laparotomy, and of the respective hemi diaphragm for patients undergoing emergency thoracotomy, in all cases of blunt or penetrating truncal trauma. If not
promptly diagnosed, TDR is associated with a high morbidity and mortality peri-operatively and in the long term. References [1] Reber PU, Schmied B, Seiler CA, Baer HU, Patel AG, Buchler MW. Missed diaphragmatic injuries and their long-term sequelae. J Trauma 1998;44(1):183–8. [2] Shah R, Sabanathan S, Mearns AJ, Choudhury AK. Traumatic rupture of diaphragm. Ann Thorac Surg 1995;60:1444–9. [3] Mansour KA. Trauma to the diaphragm. Chest Surg Clin North Am 1997;7(2):373–83. [4] Boulanger BR, Milzman DP, Rosati C, Rodriguez A. A comparison of right and left blunt traumatic diaphragmatic rupture. J Trauma 1993;35(2):255–60. [5] Murray JG, Caoili E, Gruden JF, Evans SJJ, Halvorsen Jr RA, Mackersie RC. Acute rupture of the diaphragm due to blunt trauma: diagnostic sensitivity and specificity of CT. AJR 1996;166(5):1035– 9. [6] Guth AA, Pachter HL, Kim U. Pitfalls in the diagnosis of blunt diaphragmatic injury. Am J Surg 1995;170:5–9. [7] Spann JC, Nwariaku FE, Wait M. Evaluation of video-assisted thoracoscopic surgery in the diagnosis of diaphragmatic injuries. Am J Surg 1995;170:628–31. [8] Rosati C. Acute traumatic injury of the diaphragm. Chest Surg Clin North Am 1998;8(2):371–9. [9] Johnson CD. Blunt injuries of the diaphragm. Br J Surg 1988;75:226– 30. [10] McElwee TB, Myers RT, Penell TC. Diaphragmatic rupture from blunt trauma. Am Surg 1984;50:143–9. [11] Sharma OP. Traumatic diaphragmatic rupture: not an uncommon entity—personal experience with collective review of the 1980’s. J Trauma 1989;29:678–82. [12] Leppaniemi A, Pohjankyro A, Haapiainen R. Acute diaphragmatic rupture after blunt trauma. Ann Chir Gynaecol 1994;83(1):17–21. [13] Simpson J, Lobo DN, Shah AB, Rowlands BJ. Traumatic diaphragmatic rupture: associated injuries and outcome. Ann R Coll Surg Engl 2000;82:97–100. [14] Shanmuganathan K, Mirvis SE, White CS, Pomerantz SM. MR Imaging evaluation of hemi diaphragms in acute blunt trauma: experience with 16 patients. AJR 1996;167:397–402. [15] Sarna S, Kivioja A. Blunt rupture of the diaphragm. A retrospective analysis of 41 patients. Ann Chir Gynaecol 1995;84(3):261–5. [16] Killeen KL, Mirvis SE, Shanmuganathan K. Helical CT of diaphragmatic rupture caused by blunt trauma. AJR 1999;173(6):1611–6.
Traumatic rupture of the diaphragm: experience with 65 patients Peter Mihos a , Konstantinos Potaris a,∗ , John Gakidis a , John Paraskevopoulos b , Panagiotis Varvatsoulis b , Basil Gougoutas c , George Papadakis b , Eleftherios Lapidakis c a Department of Thoracic Surgery, General Hospital of Attica ‘KAT’, 2 Nikis Street, Kifissia 14561, Greece First Department of General Surgery, General Hospital of Attica ‘KAT’, 2 Nikis Street, Kifissia 14561, Greece Second Department of General Surgery, General Hospital of Attica ‘KAT’, 2 Nikis Street, Kifissia 14561, Greece
b c
Accepted 18 October 2002
Abstract Traumatic diaphragmatic rupture is reported with increasing frequency and is associated with high morbidity and mortality. The purpose of this study was to present our experience with the management of this injury. Sixty-five patients with TDR were treated in our hospital between January 1989 and May 2000. They were 54 men (83%) and 11 women (17%). Mean age was 36.57 years (range 15–76 years). Rupture of the diaphragm was left-sided in 43 patients (66%), right-sided in 21 (32%), and bilateral in 1 (1.5%). Blunt trauma accounted for the injuries of 52 patients (80%). Early diagnosis was obtained in 57 patients (88%). The diagnosis was established preoperatively in 17 patients (26%), and intra-operatively in 48 (74%). Multiple associated injuries were observed in 62 patients (95%). Postoperative complications were observed in seven patients (11%), and the overall mortality was 14%. Injury severity score (ISS) and haemorrhagic shock upon admission strongly influence the outcome. A high index of suspicion and a thorough examination of both hemi diaphragms during laparotomy is recommended in order to avoid early or late complications. © 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction Sennertus was the first to describe a traumatic diaphragmatic hernia in 1541. Ambroise Pare diagnosed the first diaphragmatic rupture during autopsy, in 1579 [1,2]. Bowditch made an ante mortem diagnosis of a traumatic diaphragmatic rupture (TDR) in 1853 [3]. The first collective review of traumatic diaphragmatic rupture was published in 1951 and numerous studies have discussed traumatic diaphragmatic rupture since that time [4]. Acute diaphragmatic rupture occurs in 1–7% of major blunt trauma victims and in 10–15% of patients with penetrating trauma to the lower chest [1]. In conservatively managed patients the rate of initially missed diaphragmatic injuries ranges from 12 to 66%, and they may even be overlooked at laparotomy [1,5,6]. Diagnosis of a diaphragmatic injury requires a high index of suspicion. Delayed diagnosis of TDR is associated with increased morbidity and mortality and laparotomy has been suggested for all penetrating injuries of the lower thorax and upper abdomen [7]. ∗ Corresponding author. Tel.: +30-2-10-601-2587; fax: +30-2-10-231-3899. E-mail address: [email protected] (K. Potaris).
The aim of this retrospective study was to review the experience of our hospital with the management of TDR in order to identify incidence, associated morbidity and mortality, predictors of outcome, and factors contributing to diagnostic delay.
2. Material and methods The ‘KAT’ District General Hospital has a trauma center considered to be the busiest in Greece. The hospital has approximately 12,000 admissions for trauma per year. Blunt and penetrating trauma to the torso accounts for approximately 3000 admissions per year. We reviewed the notes and radiographs of 65 patients with a diagnosis of TDR treated between January 1989 and May 2000. The incidence, side of TDR, mechanism of injury, associated morbidity and mortality, and predictors of outcome were retrospectively evaluated. The majority of these patients who sustained major associated injuries were motor vehicle occupants. The variables studied and correlated with outcome included patient age, haemodynamic status upon admission, injury severity score (ISS), and time to diagnosis.
0020-1383/03/$ – see front matter © 2003 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 0 2 ) 0 0 3 6 9 - 8
170
P. Mihos et al. / Injury, Int. J. Care Injured 34 (2003) 169–172
Statistical analysis: continuous variables were compared using the unpaired Student’s t-test and differences between categorical variables were assessed using the χ 2 -test. A P-value of <0.05 was considered statistically significant.
3. Results During the specified period, 65 patients were diagnosed with TDR. Estimated incidence was 0.2% of blunt and penetrating trauma to the torso. The study identified 54 men (83%) and 11 women (17%) ranging in age from 15 to 76 years (mean 36.6 years). Causes of injury included traffic accidents (n = 51, 78%), gunshot wounds (n = 7, 11%), stab wounds (n = 6, 9%), and a fall from height (n = 1, 1.5%). TDR was left-sided in 43 patients (66%), right-sided in 21 (32%), and bilateral in 1 (1.5%). The majority of road accident induced TDR was left-sided (36 out of 51 patients, 71%). Gunshot wounds injured the right hemi diaphragm in four out of seven patients (57%). Stab wounds injured the left hemi diaphragm in four out of six patients (66.67%). Detailed data on the cause of injury and the location of TDR are given in Table 1. Diagnosis of TDR was established in less than 24 h in 57 patients (88%). In the remaining eight patients diagnostic delay ranged from 1 to 5 days with the exception of one patient who developed intra-thoracic herniation of abdominal viscera 8 years later. In 48 patients, TDR was diagnosed intra-operatively (74%) and in 17 preoperatively (26%). Diagnostic methods included chest X-ray, computerized tomography (CT) scan of the chest and upper abdomen, and upper gastrointestinal (GI) contrast study. The drainage of gastric contents via a chest tube disclosed a left TDR and a presumed iatrogenic rupture of the fundus of the stomach in one of our patients. They were both repaired during a revision laparotomy. He had undergone decortication for empyema developed after splenectomy and suturing of liver lacerations caused by a road traffic accident. 3.1. Diagnostic methods Plain chest X-rays were diagnostic of TDR in only 2 out of 65 patients (3%) but indicative of TDR in 24 patients (37%). Table 1 Cause of injury and location of traumatic diaphragmatic rupture sustained in 65 patients No. of patients
RH
LH
Bilateral
TA Falls Gunshots Stab wounds
51 1 7 6
15 (29%) 0 4 (57.1%) 2 (33.3%)
36 (71%) 1 2 (28.6%) 4 (66.7%)
0 0 1 (14.28%) 0
Total
65
21 (32.3%)
43 (66.1%)
1 (1.5%)
TA: traffic accidents; RH: right hemi diaphragm; LH: left hemi diaphragm.
They underwent either computerized tomography scan of the chest and upper abdomen or gastrografin study of the upper gastrointestinal tract, following nasogastric tube insertion. The decision depended upon the experience of the on call surgeons. CT scan of the chest and upper abdomen was performed in 10 haemodynamically stable patients. It was diagnostic of TDR in eight of them and indicative of abdominal organ herniation into the chest in the remaining two patients. Upper GI contrast study following insertion of a nasogastric tube and injection of gastrografin was performed in 14 patients. It was diagnostic of TDR in seven patients and false negative in the remainder. Diagnostic peritoneal lavage (DPL) was performed in 55 patients suspected of intra-abdominal haemorrhage and was positive for bleeding in 48 of them (87%). An intra-operative diagnosis of TDR was made in these 48 patients who underwent emergency laparotomy. DPL was indirectly diagnostic of TDR in one patient (1.5%). He had undergone chest tube insertion for haemothorax and the lavage fluid was drained through the chest tube. 3.2. Surgery The majority of our patients (n = 62, 95.4%) had multiple significant associated injuries, which are shown in Table 2. Except for the patients who had sustained both liver and spleen lacerations we observed that patients who had sustained right-sided TDR had liver injury and patients who had sustained left-sided TDR had spleen injury. Haemorrhagic shock occurred in 22 patients. Sixteen of these patients survived the operation while six died in the operating room because of profuse bleeding. Three others died later with adult respiratory distress syndrome (ARDS) in the intensive care unit (ICU). Blunt injury accounted for five of the six deaths. One patient died due to profuse bleeding, following Table 2 Multiple associated injuries observed in 62 patients sustained traumatic diaphragmatic rupture Type of injury
No. of patients
Head injury Lung laceration Pneumothorax Haemothorax Pericardial laceration Rib fractures Extremity fractures IVC laceration Spleen laceration Liver laceration Bowel laceration Kidney rupture Stomach rupture Gallbladder rupture Ovary avulsion
4 9 3 16 1 15 20 3 26 24 14 3 7 2 1
IVC: inferior vena cava.
P. Mihos et al. / Injury, Int. J. Care Injured 34 (2003) 169–172
171
Table 3 Risk factors influenced the outcome in 65 patients sustained traumatic diaphragmatic rupture
Age (years) Haemorrhagic shock (no. of patients) Early diagnosis (<24 h) (no. of patients) Late diagnosis (>24 h) (no. of patients) ISS
Survivors (n = 56)
Fatalities (n = 9)
P-value
Mean: 37.07, range: 15–68 16 (28.6%) 49 (87.5%) 7 (12.5%) Mean: 17.67, range: 9–32
Mean: 32.89, range: 29–72 6 (66.7%) 8 (88.9%) 1 (11.1%) Mean: 41.33, range: 29–72
0.330 (NS) 0.0247 (<0.05) 0.91 (NS) 0.90 (NS) 0.0004 (<0.001)
ISS: injury severity score; NS: not significant.
penetrating injury caused by a lead shot. He had TDR with associated liver and inferior vena cava lacerations. A similar but not lethal injury caused a bilateral diaphragmatic rupture to another patient who sustained major liver lacerations but no heart or great vessel injury. Two other patients underwent emergency thoracotomy because of haemothorax caused by laceration of the internal mammary artery in one, and the subclavian artery in the other. The mammary artery was ligated and the subclavian artery was primarily repaired with a polypropylene suture. They were both successfully resuscitated. Herniation of abdominal contents was observed in eight patients (12%) and included stomach in three, stomach and small bowel in one, stomach, colon and omentum in one, left colon flexure, small bowel and spleen in one, left colon flexure, left lobe of liver, spleen and upper stomach in one, and upper stomach, left colon flexure and spleen in one. The diaphragm was repaired via laparotomy in 39 patients (60%), thoracotomy in 13 (20%), and laparotomy with thoracotomy in 13 (20%). The choice depended on the clinical and radiological scenario and the associated injuries. Both interrupted and running techniques with non-absorbable suture were used. No prosthetic material was needed to close the defects. In all but one patient, surgery was performed for an acute diaphragmatic injury. Surgical treatment of the associated injuries was performed accordingly. 3.3. Outcome Postoperative complications were encountered among seven patients (11%). These included thoracic empyema and gastropleural fistula in one patient, thoracic empyema and paralytic ileus in one, mechanical ileus in one, paralytic ileus in one and pneumonia in three. They were all treated successfully either surgically or conservatively. Nine patients died (14%), six in the operating room due to non-reversible haemorrhagic shock, and the remaining three died from ARDS in the ICU. Mean hospital stay was 19.22 days (range 1–112 days). Twenty-three patients (37%) were treated in ICU. Their ICU stay ranged from 1 to 54 days with a mean of 7.3 days. Follow-up ranged from 1 to 45 months. Thirty-five of the survivors were lost to follow-up (62.5%) either because they were treated in different departments or because they failed follow-up. The remaining 21 patients had no further sequelae.
3.4. Risk factors In order to investigate risk factors influencing the outcome, we compared age, haemodynamic status, early (<24 h) or late diagnosis (>24 h), and ISS. We observed a statistically significant risk associated with haemorrhagic shock (P < 0.05), and a very significant statistically risk associated with ISS (P < 0.001). Neither age nor delay in diagnosis posed a statistically significant risk to the outcome of our patients. Observations and statistics are outlined in Table 3.
4. Comment Traumatic diaphragmatic rupture results from blunt or penetrating thoraco-abdominal trauma. It is common and regarded as a marker of trauma severity [1]. A large collective review, in 1995, suggested that 75% of the injuries to the diaphragm are caused by blunt trauma and 25% by penetrating trauma [2]. The incidence of TDR is reported to be between 0.8 and 7% when associated with blunt trauma, and between 10 and 15% when associated with penetrating trauma [8]. In a literature review in 1988, it was reported that in North American series blunt trauma accounts for 10–30% of traumatic diaphragmatic ruptures, whereas in Western Europe series blunt trauma accounts for 80–100% of TDR [9]. The true incidence of TDR is unknown because in 7–66% of major trauma victims, the diagnosis is missed. This is particularly true for ruptures of the right hemi diaphragm [1,2]. In our series, the incidence of TDR after blunt and penetrating truncal trauma was 0.2%. The principle cause of TDR was blunt trauma following road accident (77%). In blunt truncal trauma, most ruptures occur on the left posterolateral aspect of the diaphragm. This is a structurally weak area as it originates from the pleuroperitoneal membrane. The right diaphragm is congenitally stronger than the left and is partially protected by the liver, which can dissipate pressure over a large area. There is a preponderance of left-sided diaphragmatic ruptures following blunt trauma reported to be between 68.5 and 87% [2,4,10–13]. In our series, left-sided TDR occurred in 43 patients (66%). Blunt trauma accounted for the ruptures in 37 out of these 43 patients (86%).
172
P. Mihos et al. / Injury, Int. J. Care Injured 34 (2003) 169–172
There is no doubt in the literature as regards the need for aggressive operative treatment of TDR. This is due to early and late associated morbidity and mortality of this injury. We obtained an early diagnosis (less than 24 h) of TDR in 57 patients (88%). We noted 80% sensitivity of the CT scan of the chest and upper abdomen and 50% sensitivity for the upper GI contrast study. Plain chest X-ray was indicative of TDR in 24 patients (37%). Diagnostic methods that have been reported to be useful in the evaluation of TDR include plain chest X-ray, upper GI contrast study, DPL, fluoroscopic evaluation of diaphragmatic motion, ultrasound, CT scan, MRI, intra-peritoneal injection of radioisotopes, laparoscopy and video-assisted thoracic surgery. However, there are no studies to compare sensitivity, specificity and diagnostic accuracy of these modalities in the evaluation of TDR. The two modalities most commonly used, chest X-ray and CT scan are diagnostic in 30–50% of the cases. It has been reported that only 25–50% of the initial chest X-rays are diagnostic of TDR [1,5,7,8,14]. Nevertheless, in a study from Finland, chest X-rays on admission supported the diagnosis in 70% of cases and the authors concluded that a chest X-ray upon admission is the best diagnostic aid in the evaluation of TDR [15]. Sensitivity and specificity for CT scan ranges between 33–83 and 76–100%, respectively. Helical CT scanning with axial, sagittal and coronal reformations is reported to reach a sensitivity of 50% and 78% in diagnosing right- and left-sided diaphragmatic injuries, respectively [5,8,14,16]. CT scan is probably the diagnostic modality of choice in a resuscitated stable patient. DPL is reported to have a sensitivity of 66–75% and this probably relates to the degree of the associated injuries rather than diagnostic accuracy [1,7]. It was indirectly diagnostic for TDR in one of our patients (1.5%). In conclusion, an early diagnosis of TDR can easily be missed in the acute trauma setting despite the technological equipment and available diagnostic modalities. Chest X-rays or CT scans may be indicative. A thorough examination of both hemi diaphragms is recommended for all patients undergoing emergency laparotomy, and of the respective hemi diaphragm for patients undergoing emergency thoracotomy, in all cases of blunt or penetrating truncal trauma. If not
promptly diagnosed, TDR is associated with a high morbidity and mortality peri-operatively and in the long term. References [1] Reber PU, Schmied B, Seiler CA, Baer HU, Patel AG, Buchler MW. Missed diaphragmatic injuries and their long-term sequelae. J Trauma 1998;44(1):183–8. [2] Shah R, Sabanathan S, Mearns AJ, Choudhury AK. Traumatic rupture of diaphragm. Ann Thorac Surg 1995;60:1444–9. [3] Mansour KA. Trauma to the diaphragm. Chest Surg Clin North Am 1997;7(2):373–83. [4] Boulanger BR, Milzman DP, Rosati C, Rodriguez A. A comparison of right and left blunt traumatic diaphragmatic rupture. J Trauma 1993;35(2):255–60. [5] Murray JG, Caoili E, Gruden JF, Evans SJJ, Halvorsen Jr RA, Mackersie RC. Acute rupture of the diaphragm due to blunt trauma: diagnostic sensitivity and specificity of CT. AJR 1996;166(5):1035– 9. [6] Guth AA, Pachter HL, Kim U. Pitfalls in the diagnosis of blunt diaphragmatic injury. Am J Surg 1995;170:5–9. [7] Spann JC, Nwariaku FE, Wait M. Evaluation of video-assisted thoracoscopic surgery in the diagnosis of diaphragmatic injuries. Am J Surg 1995;170:628–31. [8] Rosati C. Acute traumatic injury of the diaphragm. Chest Surg Clin North Am 1998;8(2):371–9. [9] Johnson CD. Blunt injuries of the diaphragm. Br J Surg 1988;75:226– 30. [10] McElwee TB, Myers RT, Penell TC. Diaphragmatic rupture from blunt trauma. Am Surg 1984;50:143–9. [11] Sharma OP. Traumatic diaphragmatic rupture: not an uncommon entity—personal experience with collective review of the 1980’s. J Trauma 1989;29:678–82. [12] Leppaniemi A, Pohjankyro A, Haapiainen R. Acute diaphragmatic rupture after blunt trauma. Ann Chir Gynaecol 1994;83(1):17–21. [13] Simpson J, Lobo DN, Shah AB, Rowlands BJ. Traumatic diaphragmatic rupture: associated injuries and outcome. Ann R Coll Surg Engl 2000;82:97–100. [14] Shanmuganathan K, Mirvis SE, White CS, Pomerantz SM. MR Imaging evaluation of hemi diaphragms in acute blunt trauma: experience with 16 patients. AJR 1996;167:397–402. [15] Sarna S, Kivioja A. Blunt rupture of the diaphragm. A retrospective analysis of 41 patients. Ann Chir Gynaecol 1995;84(3):261–5. [16] Killeen KL, Mirvis SE, Shanmuganathan K. Helical CT of diaphragmatic rupture caused by blunt trauma. AJR 1999;173(6):1611–6.