TY - JOUR
AU1 - Gooseman, Michael, R
AU2 - Rawashdeh,, Majd
AU3 - Mattam,, Kavitha
AU4 - Rao, Jagan, N
AU5 - Vaughan, Paul, R
AU6 - Edwards, John, G
AB - Abstract View largeDownload slide View largeDownload slide OBJECTIVES Taxonomy of injuries involving the costal margin is poorly described and surgical management varies. These injuries, though commonly caused by trauma, may also occur spontaneously, in association with coughing or sneezing, and can be severe. Our goal was to describe our experience using sequential segmental analysis of computed tomographic (CT) scans to perform accurate assessment of injuries around the costal margin. We propose a unifying classification for transdiaphragmatic intercostal hernia and other injuries involving the costal margin. We identify the essential components and favoured techniques of surgical repair. METHODS Patients presenting with injuries to the diaphragm or to the costal margin or with chest wall herniation were included in the study. We performed sequential segmental analysis of CT scans, assessing individual injury patterns to the costal margin, diaphragm and intercostal muscles, to create 7 distinct logical categories of injuries. Management was tailored to each category, adapted to the individual case when required. Patients with simple traumatic diaphragmatic rupture were considered separately, to allow an estimation of the relative incidence of injuries to the costal margin compared to those of the diaphragm alone. RESULTS We identified 38 patients. Of these, 19 had injuries involving the costal margin and/or intercostal muscles (group 1). Sixteen patients in group 1 underwent surgery, 2 of whom had undergone prior surgery, with 4 requiring a novel double-layer mesh technique. Nineteen patients (group 2) with diaphragmatic rupture alone had a standard repair. CONCLUSIONS Sequential analysis of CT scans of the costal margin, diaphragm and intercostal muscles defines accurately the categories of injury. We propose a ‘Sheffield classification’ in order to guide the clinical team to the most appropriate surgical repair. A variety of surgical techniques may be required, including a single- or double-layer mesh reinforcement and plate and screw fixation. Costal margin, Transdiaphragmatic intercostal hernia, Diaphragm, Intercostal hernia INTRODUCTION Injuries involving the costal margin, with or without diaphragmatic or intercostal muscle injuries, are rare, and diagnosis and management are challenging. Costal margin rupture (CMR) was described by Roland in 1499 [1], and ‘intercostal diaphragmatic hernia’ in the context of trauma was described by Gerster in 1911 [2]. Aetiology includes blunt trauma, fall, crush, deceleration and seat belt injuries. Apparently spontaneous injuries, occurring with coughing or sneezing, may also occur [3–5]. Penetrating trauma and iatrogenic injuries are less common. Despite its long-standing recognition, the literature comprises only case reports and small series, with variable (or absent) descriptions of CMR and inconsistent descriptions of injuries to the associated muscles, reported variably as transdiaphragmatic intercostal hernia, intercostal hernia or acquired abdominal intercostal hernia, for example [3–19]. However, a review of the literature revealed that the majority of reported cases display CMR, unmentioned in many cases, with or without the presence of injury to the diaphragm or intercostal muscles (Supplementary Material, Table S1). The intraoperative technique tends to be poorly described, and postoperative outcomes are universally favourable. Hence, both nomenclature and surgical technique are variable. Following identification of our first case [3], we noted a steady incidence of similar injuries. Our goal was (i) to review our experience of costal margin injuries; (ii) to derive a systematic injury classification using sequential segmental analysis similar to that used in describing complex congenital cardiac abnormalities [20]; and (iii) to provide practical recommendations for the management of patients with injuries involving the costal margin. We also describe 2 novel techniques: a 5-layer double-mesh repair and the use of titanium plates and screws, which we have found effective in challenging cases. Our hypothesis is that sequential segmental analysis of computed tomographic (CT) scans allows the most accurate characterization of the injury, which guides the appropriate management plan. METHODS Patients We identified prospectively patients presenting between January 2006 and April 2017 with CMR, with or without diaphragmatic rupture (DR) or intercostal herniation (IH) (group 1). Patients with isolated DR but without CMR were included (group 2) to allow an estimation of the relative incidence of CMR. However, characterization of group 2 is not the primary focus of the study because the management of DR is well described and not controversial. Furthermore, we included only patients with DR managed by the thoracic surgeons in our unit, because some patients with DR in our unit were managed by general surgeons alone. Patients were cared for in a general thoracic surgery unit, within a major trauma centre serving a population of 1.8 million. We applied systematic sequential segmental analysis to axial CT images, with coronal and sagittal planar reconstruction where necessary, allowing characterization and categorization of injuries to the costal margin, diaphragm and intercostal muscles in each case (Fig. 1). Figure 1: View largeDownload slide Sheffield classification of injuries involving the costal margin. CMR: costal margin rupture; DR: diaphragmatic rupture; IH: intercostal hernia; TDIH: traumatic diaphragmatic/intercostal hernia. Figure 1: View largeDownload slide Sheffield classification of injuries involving the costal margin. CMR: costal margin rupture; DR: diaphragmatic rupture; IH: intercostal hernia; TDIH: traumatic diaphragmatic/intercostal hernia. Surgical techniques All patients undergoing surgery under our care received general anaesthesia and single-lung ventilation via double-lumen endotracheal tube intubation. Patients were positioned laterally or supinely to allow the best surgical exposure of the injury. The chest cavity was accessed by incision over the CMR and IH (where present), via a thoracotomy or thoraco-laparotomy, as appropriate. Full intraoperative assessment of the costal margin, diaphragm and intercostal muscles was undertaken to confirm the CT images, and subsequent surgical procedures were tailored according to the assigned injury category: IH: external mesh repair, using a polypropylene or biological mesh secured with non-absorbable sutures. DR: primary suture repair of the diaphragm, with interrupted Teflon (Ethicon, Somerville, NJ, USA) pledgeted or continuous Ethibond (Ethicon) sutures. CMR: conservative management was used initially, although surgery, by 1 of 2 techniques, was offered to patients with persistent symptomatic non-union. Formerly, 2 No. 5 Ethibond sutures were used to reapproximate the CMR; an additional 3 pericostal No. 2 Ethibond sutures were placed, laterally reapproximating the intercostal space (IS). Laterally, external titanium plates with locking titanium screws (MatrixRib, Synthes, West Chester, PA, USA) were used [3] (Fig. 2). Although it is preferable to span from the rib across to the sternum, this may not be possible for CMR at lower levels. In such cases, fixation of titanium plates to costal cartilage alone has been used, with 3 or 4 locking screws placed on either side of the CMR after using No. 5 Ethibond sutures to provide apposition of the cartilage ends. CMR + IH: a combination of the foregoing techniques, with sutures to the costal margin and either an external mesh or double-layer mesh repair. Fixation of symptomatic non-united rib fractures may be performed. CMR + DR: primary repair with sutures, combining the described previously techniques for CMR and DR. Traumatic diaphragmatic/intercostal hernia (TDIH): We found that suture-only repair was associated with failure. Therefore, we developed a ‘5-layer double-mesh repair’. The origins of the serratus anterior and external oblique muscles were mobilized from the chest wall, exposing the ribs 3 spaces below and 2 spaces above the defect. Two layers of polypropylene mesh (Ethicon) were secured, 1 inside and 1 outside the chest, with No. 1 Prolene sutures (Ethicon), covering the zone from 2 spaces below to 2 spaces above the intercostal defect. Two 2-cm × 1-cm Teflon felt (Ethicon) rectangular pledgets were cut for each horizontal mattress-securing suture. Sutures were placed from the outside, passing through the Teflon pledget, the outer mesh, the intercostal space, the inner mesh and the inner Teflon pledget, returning back through those 5 layers in reverse. Sutures were placed in the 2 spaces caudal to the rupture first and then secured (Fig. 3). The sutures cephalad to the defect were placed with the temporary aid of a Tudor-Edwards rib approximator (GU Medical Ltd, Reading, UK), marking accurately the cephalad extent of each suture through the meshes, but these sutures were not yet tied. Sutures were passed through snugger tubes, to check the suture position under tension. The diaphragm was repaired after opening the peritoneal membrane (if required), to allow safe placement, 1 cm back from the ruptured edge of the diaphragm, of No. 2 Ethibond horizontal continuous mattress sutures, which were then oversewn with a second layer. Transverse abdominis and external oblique muscles were repaired with continuous No. 2 Ethibond sutures. The CMR was repaired with sutures, as described previously. Conventional pericostal sutures were placed, passing sutures through holes punched into the lower rib bordering the defect with a Sweet’s sternal punch (GU Medical Ltd, Reading, UK), to protect the intercostal bundle, and were tied. In all cases, the area was washed out with a dilute solution of iodine/povidone before placement of a 28 Fr intercostal tube drain. After tying the diaphragm, abdominal wall and intercostal sutures, the mesh repair sutures superior to the defect were secured (Fig. 4, Supplementary Material, Figs S1–S17). Muscle and skin layers were closed with Vicryl (Ethicon, US), and Monocryl (Ethicon, US) sutures. Small-bore suction drains (Redivac, Iseki, Daventry, UK) were placed deep to the muscle layers. The chest drain was connected to an underwater drainage system (Sentinel Seal, Medtronic, Minneapolis, MN, USA) with the application of −20 cmH20 negative suction. Patients were extubated on the operating table whenever possible. There was routine engagement with an enhanced recovery after surgery programme, encouraging early nutrition, mobilization, physical therapy and drain removal. Figure 2: View largeDownload slide Chest radiograph of a case of painful, non-united costal margin rupture after plate and screw fixation. Figure 2: View largeDownload slide Chest radiograph of a case of painful, non-united costal margin rupture after plate and screw fixation. Figure 3: View largeDownload slide Repair of traumatic diaphragmatic/intercostal hernia (left lateral position, posterior aspect). Extrathoracic and intrathoracic polypropylene meshes are seen on either side of the rib, forming the caudal border of the intercostal herniation. Figure 3: View largeDownload slide Repair of traumatic diaphragmatic/intercostal hernia (left lateral position, posterior aspect). Extrathoracic and intrathoracic polypropylene meshes are seen on either side of the rib, forming the caudal border of the intercostal herniation. Figure 4: View largeDownload slide Repair of traumatic diaphragmatic/intercostal hernia. All sutures are tied. There are rows of 4 sutures, both 1 and 2 intercostal spaces caudal and cephalad to the intercostal herniation. Figure 4: View largeDownload slide Repair of traumatic diaphragmatic/intercostal hernia. All sutures are tied. There are rows of 4 sutures, both 1 and 2 intercostal spaces caudal and cephalad to the intercostal herniation. As part of the literature review, we have undertaken for each reported case we analysed (where possible) the description of the injury and categorized each according to our proposed classification (Supplementary Material, Table S1). RESULTS Thirty-eight patients were included (Fig. 1). There were 19 patients in group 1 (injuries involving the costal margin or intercostal muscles). There were 19 cases with DR alone (group 2). In group 1, 18 (95%) were men; the median age was 59 years, and the body mass index was 30 kg/m2 (Table 1). There was a positive correlation between a body mass index greater than 30 and CMR compared to DR alone (P = 0.003, χ2 test). Eleven (58%) of the 19 patients in group 1 suffered the injury following a cough or sneeze (Table 2) compared to none with DR alone (group 2). There were no consistent underlying medical conditions in the patients with cough-induced injury, such as chronic obstructive pulmonary disease, diabetes mellitus or steroid use. CMR was reported by the radiologist in 12 (67%) of 18 cases where it occurred. In 6 cases the CMR was at the level of the 6th intercostal space, 3 at the 7th, 7 at the 8th and 2 at the 9th intercostal space. There was no correlation between the aetiology and the level of CMR. Figure 5 illustrates CT images of TDIH; Video 1 shows the CT scan of CMR with IH, and Video 2 shows TDIH. Figure 5: View largeDownload slide Computed tomographic scan of a case of recurrent traumatic diaphragmatic/intercostal hernia after a high-velocity motor vehicle collision, as described in the text. Representative axial, sagittal and coronal sections are shown in addition to the 3-dimensional reconstructions, which demonstrate non-united posterior sector 7th and 8th rib fractures, with the CMR component present in the 8th intercostal space. An aortic stent, placed for traumatic aortic injury, is also seen. Figure 5: View largeDownload slide Computed tomographic scan of a case of recurrent traumatic diaphragmatic/intercostal hernia after a high-velocity motor vehicle collision, as described in the text. Representative axial, sagittal and coronal sections are shown in addition to the 3-dimensional reconstructions, which demonstrate non-united posterior sector 7th and 8th rib fractures, with the CMR component present in the 8th intercostal space. An aortic stent, placed for traumatic aortic injury, is also seen. Video 1: Coronal computed tomographic images of a case of costal margin rupture + intercostal herniation. There is a costal margin rupture between the right 8th and 9th ribs and an intercostal hernia revealed by splaying of the ribs, but the diaphragm remains intact. Video 1: Coronal computed tomographic images of a case of costal margin rupture + intercostal herniation. There is a costal margin rupture between the right 8th and 9th ribs and an intercostal hernia revealed by splaying of the ribs, but the diaphragm remains intact. Close Video 2: Axial computed tomographic images of a case of a traumatic diaphragmatic/intercostal hernia. There is a costal margin rupture between the left 9th and 10th ribs, with associated intercostal herniation and diaphragm rupture. Video 2: Axial computed tomographic images of a case of a traumatic diaphragmatic/intercostal hernia. There is a costal margin rupture between the left 9th and 10th ribs, with associated intercostal herniation and diaphragm rupture. Close Table 1: Patient demographics Group 1 (injuries involving the costal margin) (n = 19), n (%) Group 2 (isolated diaphragmatic rupture) (n = 19), n (%) Gender  Male 18 (95) 14 (74)  Female 1 (5) 5 (26) Age (years)  ≤64 12 (63) 13 (68)  65–74 6 (32) 5 (26)  ≥75 1 (5) 1 (5) Body mass index (kg/m2)  ≤17 0 (0) 1 (5)  18–23 1 (5) 4 (21)  24–30 6 (32) 10 (53)  ≥30 12 (63) 3 (21) Group 1 (injuries involving the costal margin) (n = 19), n (%) Group 2 (isolated diaphragmatic rupture) (n = 19), n (%) Gender  Male 18 (95) 14 (74)  Female 1 (5) 5 (26) Age (years)  ≤64 12 (63) 13 (68)  65–74 6 (32) 5 (26)  ≥75 1 (5) 1 (5) Body mass index (kg/m2)  ≤17 0 (0) 1 (5)  18–23 1 (5) 4 (21)  24–30 6 (32) 10 (53)  ≥30 12 (63) 3 (21) View Large Table 1: Patient demographics Group 1 (injuries involving the costal margin) (n = 19), n (%) Group 2 (isolated diaphragmatic rupture) (n = 19), n (%) Gender  Male 18 (95) 14 (74)  Female 1 (5) 5 (26) Age (years)  ≤64 12 (63) 13 (68)  65–74 6 (32) 5 (26)  ≥75 1 (5) 1 (5) Body mass index (kg/m2)  ≤17 0 (0) 1 (5)  18–23 1 (5) 4 (21)  24–30 6 (32) 10 (53)  ≥30 12 (63) 3 (21) Group 1 (injuries involving the costal margin) (n = 19), n (%) Group 2 (isolated diaphragmatic rupture) (n = 19), n (%) Gender  Male 18 (95) 14 (74)  Female 1 (5) 5 (26) Age (years)  ≤64 12 (63) 13 (68)  65–74 6 (32) 5 (26)  ≥75 1 (5) 1 (5) Body mass index (kg/m2)  ≤17 0 (0) 1 (5)  18–23 1 (5) 4 (21)  24–30 6 (32) 10 (53)  ≥30 12 (63) 3 (21) View Large Table 2: Categorization, presentation and aetiology of injuries Study group Full name (Abbreviation) Number of patients Presentation Cause Acute Chronic High velocity trauma Fall Cough or sneeze Crush injury Group 1 Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 0 1 0 3 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 0 3 0 2 0 Costal margin rupture with intercostal hernia (CMR + IH) 4 3 1 0 0 4 0 Costal margin rupture (CMR) 5 4 1 1 1 2 1 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 0 0 0 0 0 0 Intercostal hernia (IH) 1 1 0 0 1 0 0 Group 2 Diaphragmatic rupture (DR) 19 14 5 19 0 0 0 Study group Full name (Abbreviation) Number of patients Presentation Cause Acute Chronic High velocity trauma Fall Cough or sneeze Crush injury Group 1 Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 0 1 0 3 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 0 3 0 2 0 Costal margin rupture with intercostal hernia (CMR + IH) 4 3 1 0 0 4 0 Costal margin rupture (CMR) 5 4 1 1 1 2 1 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 0 0 0 0 0 0 Intercostal hernia (IH) 1 1 0 0 1 0 0 Group 2 Diaphragmatic rupture (DR) 19 14 5 19 0 0 0 View Large Table 2: Categorization, presentation and aetiology of injuries Study group Full name (Abbreviation) Number of patients Presentation Cause Acute Chronic High velocity trauma Fall Cough or sneeze Crush injury Group 1 Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 0 1 0 3 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 0 3 0 2 0 Costal margin rupture with intercostal hernia (CMR + IH) 4 3 1 0 0 4 0 Costal margin rupture (CMR) 5 4 1 1 1 2 1 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 0 0 0 0 0 0 Intercostal hernia (IH) 1 1 0 0 1 0 0 Group 2 Diaphragmatic rupture (DR) 19 14 5 19 0 0 0 Study group Full name (Abbreviation) Number of patients Presentation Cause Acute Chronic High velocity trauma Fall Cough or sneeze Crush injury Group 1 Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 0 1 0 3 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 0 3 0 2 0 Costal margin rupture with intercostal hernia (CMR + IH) 4 3 1 0 0 4 0 Costal margin rupture (CMR) 5 4 1 1 1 2 1 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 0 0 0 0 0 0 Intercostal hernia (IH) 1 1 0 0 1 0 0 Group 2 Diaphragmatic rupture (DR) 19 14 5 19 0 0 0 View Large At a median 32 (range 19–145) months of follow-up, 16 patients in group 1 have undergone 18 operations (Table 3). All 4 patients with TDIH underwent surgery at presentation. One had repair of the DR only by a general surgeon at the same time as an end colostomy. The TDIH recurred at 18 months and required a redo 5-layer double-mesh repair and costal margin plate fixation (Fig. 5). One of 6 patients with CMR + DR was noted to have abdominal cavity free air and underwent an urgent laparotomy and DR repair without the CMR being recognized: This patient (with multiple injuries) died of acute lung injury 3 days after admission. Of 2 other patients with CMR + DR who had suture-only repair at the time of presentation, 1 recurred, with a redo 5-layer double-mesh repair performed 8 months later. Three of the 4 patients with CMR + IH underwent urgent surgical repair with sutures only. The fourth patient was treated conservatively at his request, but he developed significant symptoms at 21 months and a 5-layer double- mesh repair was performed. Of those patients with isolated CMR, 2 received suture repair at presentation and plate fixation was carried out in 1 patient with a symptomatic non-union of the CMR. Two patients with CMR were treated conservatively. One patient with isolated IH had an extrathoracic mesh repair. We have not seen any cases of TDIH without CMR. All patients with isolated DR (group 2) received urgent surgical repair. Table 3: Operative management of injuries involving the costal margin Full name (Abbreviation) Number of patients Number requiring surgery Number of surgical procedures Extrathoracic mesh repair Five-layer double-mesh repair Suture repair Costal margin plate fixation Rib plate fixation Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 5a 1 3 1 (bdiaphragm only) 1 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 6 1 1 3b (1 diaphragm only) 2 1 Costal margin rupture with intercostal hernia (CMR + IH) 4 3c 3 0 0 3 0 1 Costal margin rupture (CMR) 5 3 3 0 0 3 1 2 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 NA NA NA NA NA NA NA Intercostal hernia (IH) 1 1 1 1 0 0 0 0 Diaphragmatic rupture (DR) 19 19 19 0 0 19 0 0 Full name (Abbreviation) Number of patients Number requiring surgery Number of surgical procedures Extrathoracic mesh repair Five-layer double-mesh repair Suture repair Costal margin plate fixation Rib plate fixation Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 5a 1 3 1 (bdiaphragm only) 1 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 6 1 1 3b (1 diaphragm only) 2 1 Costal margin rupture with intercostal hernia (CMR + IH) 4 3c 3 0 0 3 0 1 Costal margin rupture (CMR) 5 3 3 0 0 3 1 2 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 NA NA NA NA NA NA NA Intercostal hernia (IH) 1 1 1 1 0 0 0 0 Diaphragmatic rupture (DR) 19 19 19 0 0 19 0 0 a Costal margin plate fixation is being considered by 1 patient with persistent pain following 5-layer double-mesh repair. b One patient with TDIH and 1 with CMR + DR underwent suture repair at presentation but required repeat surgery with 5-layer double-mesh repair for recurrence. c One patient with CMR + IH was treated conservatively but after 21 months has significant symptoms and awaits repair. NA: not applicable. View Large Table 3: Operative management of injuries involving the costal margin Full name (Abbreviation) Number of patients Number requiring surgery Number of surgical procedures Extrathoracic mesh repair Five-layer double-mesh repair Suture repair Costal margin plate fixation Rib plate fixation Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 5a 1 3 1 (bdiaphragm only) 1 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 6 1 1 3b (1 diaphragm only) 2 1 Costal margin rupture with intercostal hernia (CMR + IH) 4 3c 3 0 0 3 0 1 Costal margin rupture (CMR) 5 3 3 0 0 3 1 2 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 NA NA NA NA NA NA NA Intercostal hernia (IH) 1 1 1 1 0 0 0 0 Diaphragmatic rupture (DR) 19 19 19 0 0 19 0 0 Full name (Abbreviation) Number of patients Number requiring surgery Number of surgical procedures Extrathoracic mesh repair Five-layer double-mesh repair Suture repair Costal margin plate fixation Rib plate fixation Traumatic diaphragmatic and intercostal hernia (TDIH) 4 4 5a 1 3 1 (bdiaphragm only) 1 0 Costal margin rupture with diaphragmatic rupture (CMR + DR) 5 5 6 1 1 3b (1 diaphragm only) 2 1 Costal margin rupture with intercostal hernia (CMR + IH) 4 3c 3 0 0 3 0 1 Costal margin rupture (CMR) 5 3 3 0 0 3 1 2 Traumatic diaphragmatic and intercostal hernia without costal margin rupture (TDIH w/o CMR) 0 NA NA NA NA NA NA NA Intercostal hernia (IH) 1 1 1 1 0 0 0 0 Diaphragmatic rupture (DR) 19 19 19 0 0 19 0 0 a Costal margin plate fixation is being considered by 1 patient with persistent pain following 5-layer double-mesh repair. b One patient with TDIH and 1 with CMR + DR underwent suture repair at presentation but required repeat surgery with 5-layer double-mesh repair for recurrence. c One patient with CMR + IH was treated conservatively but after 21 months has significant symptoms and awaits repair. NA: not applicable. View Large Of the 18 surgical procedures in the group 1, after 15 operations the patients were extubated immediately in the operating room and transferred to the recovery area. Two patients required high oxygen concentrations intraoperatively and were transferred to the intensive care unit (ICU) before extubation the following day. One additional patient with CMR + DR who underwent repair of the DR only remained intubated postoperatively until death. One patient required a 6-day stay in the ICU due to type 2 respiratory failure, without reintubation. Another patient remained in the ICU for 15 days while recovering from significant orthopaedic injuries. Among the patients undergoing surgery, at 1 year, 1 CMR + DR patient (5.3%) in group 1 died; no patients died in group 2. Although we did not prospectively record health-related quality of life using questionnaires, we assessed narrative outcomes at the post-discharge hospital visits. One patient with TDIH treated with a 5-layer double-mesh repair reported a significant burden of pain, requiring regular pregabalin (Lyrica, Pfizer, New York, NY, USA). We believe this to be related to the non-union of the CMR, for which the patient is considering whether to undergo further surgery with plate and screw fixation. All other patients reported being symptom-free at the most recent postsurgical follow-up and had returned to preinjury activities of daily living, including employment. DISCUSSION A literature review (Supplementary Material, Table S1) demonstrates inconsistent terminology where it is possible to categorize injuries according to the proposed classification. TDIH is often reported to be rare (even in recent publications), but there are many single case reports in the literature. The main reason for the lack of identification of previous cases by case report authors may be the inconsistent taxonomy. Descriptions include transdiaphragmatic intercostal hernia, intercostal hernia or acquired abdominal intercostal hernia and pseudohernia [3–19]. TDIH is frequently described by other authors as a DR and then intercostal hernia back into the chest, through an intercostal space weak due to fractures, or the absence of 2 layers of intercostal muscles anteriorly and posteriorly—but without any acknowledgement of CMR [14, 17, 18]. Although older reports did not include images of CT scans, from more recent reports it was usually possible to determine whether a CMR was present. In 26 of 39 reported cases where we judged there to be CMR, it was not mentioned in the manuscript. Hence, there is a need for a systematic classification to categorize injuries accurately, using the method we described previously. It is difficult to assign meaningful diagnostic codes for costal margin injuries using existing coding systems. There is no appropriate international classification of diseases code, for example. Although it is possible to code a costochondral fracture using the Orthopaedic Trauma Association/AO Foundation Fracture and Dislocation Classification Compendium, the associated soft tissue injuries are not included [21]. The American Association for the Surgery of Trauma Organ Injury Scales (OIS) for diaphragmatic injuries and chest wall injuries do not enable accurate classification of the spectrum of injuries we describe. The American Association for the Surgery of Trauma OIS for the diaphragm characterizes the presence of contusion, the presence and length of laceration and the area of tissues loss [22]. The OIS for the chest wall categorizes the presence of contusions, lacerations and fractures. However, although the presence of adjacent rib fractures and flail segments is included, there is no OIS category suitable for CMR that acknowledges the variety of associated soft tissue injuries that we describe [23]. Having analysed our series of injuries around the costal margin, we propose a systematic and unifying classification of injury (Fig. 1), based upon the concept of sequential segmental analysis [20]. The essence of cardiac sequential segmental analysis is the systematic evaluation of the anatomical elements and the junctions between them. Here we define the costal margin as the crucial ‘junction’, to which are anatomically related the diaphragm, intercostal muscles and anterior abdominal wall. The circumferential attachment of the diaphragm is predominantly to the costal margin but also to the upper 3 lumbar vertebra and the anterior tips of the 12th and 11th ribs. For a continuous tear to extend from the diaphragm to an intercostal space, we propose that the injury must involve a CMR. The lack of direct continuity between the diaphragm and the intercostal muscles, with these being separated by the costochondral cartilaginous margin at levels above the 10th rib, accounts for our failure thus far to identify a case of TDIH with CMR being present. We found only 1 possible case in the literature [19], but the status of the costal margin was not mentioned in that report. Furthermore, although we have encountered a case of isolated IH, we believe that it is rare to see splaying or separation of an intercostal space without the presence of displaced rib or costochondral fractures, sternocostal joint dislocation or a CMR. Similar to the published reports, our radiologists did not record the CMR in 33% of initial CT reports where we later judged it to be present. It is surprising that many previous authors also did not note the CMR. However, we found that CMR frequently evades initial detection, both in the literature and in our own experience. The reasons for this omission are not clear. Our institution is a major trauma centre, with a protocol for a radiologist reporting within 30 min of the CT scan being performed. It is hypothesized that, unless there is awareness by the radiologist of the anatomy of these relatively rare injuries, it is easy to overlook injury to the costal cartilages because of their complex shape and the fact that their density is similar to that of the surrounding soft tissues. When the precise nature of the injury is not recognized, we have found that patients may either be managed non-operatively (only to return many months or years later) or may undergo unsuccessful surgery that does not address the CMR. Two of our patients underwent surgery for presumed isolated DR: The CMR remained unrecognized. Application of our proposed methods of classification should prevent the CMR from being missed. We have considered whether the mode of injury and type of presentation may affect the appropriate method of repair. Considering that 2 of the recurrences seen were in patients in whom the extent of the injury was underestimated, we believe that our diagnostic algorithm should be followed, and plans should be made for the repair of all components. Whether this can be achieved at presentation will depend on the stability of the patient: In the setting of a damage-control laparotomy or thoracotomy, immediate repair may not be performed, but plans should be made for subsequent repair as soon as the patient is stable. Almost all of our patients were in a stable condition at presentation. We note that only cases of isolated CMR have avoided surgical management, with eventual failure of cases managed conservatively in other categories. Thus, we propose surgical repair at presentation, according to the stability of the patient, for all injury categories except isolated CMR. For isolated CMR, early surgical repair may be indicated for pain control or subsequently if there is painful non-union. We believe that it is important to address surgically each of the 4 possible components of injury around the costal margin: the diaphragm, the intercostal muscles, the abdominal wall and the costal margin itself. We have demonstrated that CMR may (or may not) lead to a tear in any 1, 2 or all 3 of the associated muscular sheets. We have demonstrated that there is a risk of recurrence with incomplete repair, particularly when the CMR is not repaired. Even with the stability gained by the 5-layer double-mesh repair, significant symptoms may remain if non-union of the CMR persists. Furthermore, we are aware, from discussions with colleagues in other thoracic surgery units, of metalwork fracture and recurrence of TDIH in cases where there was repair of the DR and surgical stabilization of associated rib fractures alone, but without fixation of the CMR or mesh repair of the intercostal hernia components. We believe that the 5-layer double-mesh repair provides the most secure solution for complex injury patterns. Having seen failure of a suture-only repair in an obese patient with cough-induced TDIH, we did not feel that a single-layer extrathoracic mesh would provide sufficient support. It is our opinion that the double-layer mesh technique optimally distributes the forces across the intercostal defect (from coughing, for example). We have not yet experienced failure of this repair. We propose an intercostal-sparing technique for future cases, where the ribs caudal to the defect are drilled for suture placement rather than using the intercostal horizontal mattress pledgeted sutures, in order to minimize the pressure on the intercostal nerves. We have safely used plate and screw fixation in the management of CMR, with fixation into the costal cartilage alone rather than into bone, using the equipment and techniques used routinely for rib fixation. Because we initiated elastic stable chest repair, we are familiar with the techniques of fixing costal cartilage to repair complex pectus deformities [24]. Experience from our units and colleagues elsewhere indicates that the placement of plate and locking screw fixation of cartilage does not result in construct or healing failure [25]. We propose that combining the 5-layer double-mesh repair with costal margin plate and screw fixation provides the most secure repair in cases of CMR + IH or TDIH. The surgical results in the literature are universally favourable. However, we have encountered patients who experience persistent postoperative pain and the need for reoperation due to recurrence of herniation (2 cases) and symptomatic CMR non-union (1 case). Hence it is important to characterize injuries thoroughly with the proposed classification in order to collect accurate outcome data and determine the appropriate method of surgical repair. Limitations This study is limited by the exclusion of some cases of DR repaired by other surgeons. Furthermore, whereas the cases with the IH component are likely to have been referred to the thoracic surgical service, there may also have been cases presenting to our trauma network with CMR + DR or CMR that were unknown to us. Because ICD-10 codes do not record CMR or the categories of injury we have derived in this study, we do not have the ability to retrieve cases from hospital coding data that we have not identified in the prospective study. The number of isolated patients with CMR may also be an underestimate of the true incidence. One of the patients in this series was managed without hospital admission. It is possible that there are patients discharged from emergency departments without a CT scan and characterization of the injury. It is a matter of conjecture, but it is possible that some of these patients might benefit from surgical management. Hence the definitive incidence of injuries associated with CMR and their precise ratio to isolated DR remain unknown. An additional limitation is that we have not included associated rib fractures, non-floating costochondral fractures or sternochondral joint disruptions in the classification. These may contribute to instability and may affect the surgical strategy. CONCLUSION In conclusion, we recommend that, for those patients who have splaying of the ribs or in whom traumatic intercostal hernia or DR is identified, there should be careful CT assessment of the costal margin to determine whether CMR is present. Although not used in this series, 3-dimensional reconstruction of CT scans with composite volume rendering of the bone and costal cartilage may help provide a more sensitive characterization of the injury. Sequential segmental analysis of the costal margin, diaphragm and intercostal muscles should be used to categorize the injuries according to the proposed classification and to guide management. Surgery, where indicated, should address each component of injury in order to minimize failure of repair. We recommend adoption of our taxonomy by emergency department staff, radiologists and surgeons to allow the rapid and systematic assessment of the injury, facilitate communication and ensure application of the most appropriate management plan. Conflict of interest: none declared. Presented at the Chest Wall Injury Society, Salt Lake City, UT, USA, 3 March 2018. REFERENCES 1 Roland. De pulmonis sanarpot, Liber III (cap) XXV. In: de Chavliae G (ed). Cyrurgia. 1499, 144. 2 Gerster JC. Intercostal diaphragmatic hernia: with report of a case . Ann Surg 1911 ; 54 : 538 – 48 . Google Scholar Crossref Search ADS PubMed 3 Waqar S , Vaughan P , Hill C , Edwards JG. Late presentation of a traumatic lung hernia . Ann Thorac Surg 2010 ; 90 : 1715. Google Scholar Crossref Search ADS PubMed 4 McAdams TR , Deimel JF , Ferguson J , Beamer BS , Beaulieu CF. Chondral rib fractures in professional american football: two cases and current practice patterns among NFL team physicians . Orthop J Sports Med 2016 ; 4 : 2 . doi: 10.1177/2325967115627623. Google Scholar Crossref Search ADS 5 Chaar CIO , Attanasio P , Detterbeck F. Disruption of the costal margin with transdiaphragmatic abdominal herniation induced by coughing . Am Surg 2008 ; 74 : 350 – 3 . Google Scholar PubMed 6 Rogers FB , Leavitt BJ , Jensen PE. Traumatic transdiaphragmatic intercostal hernia secondary to coughing: case report and review of the literature . J Trauma Acute Care Surg 1996 ; 41 : 902 – 3 . Google Scholar Crossref Search ADS 7 Daniel R , Naidu B , Khalil-Marzouk J. Cough-induced rib fracture and diaphragmatic rupture resulting in simultaneous abdominal visceral herniation into the left hemithorax and subcutaneously . Eur J Cardiothorac Surg 2008 ; 34 : 914 – 15 . Google Scholar Crossref Search ADS PubMed 8 Connery A , Mutvalli E. Cough-induced abdominal intercostal hernia . JRSM Short Rep 2010 ; 1 : 1 – 3 . Google Scholar Crossref Search ADS PubMed 9 Sit A , Chatterjee S , Das I , Datta A , Jana PK , Joseph JK et al. A case of spontaneous transdiaphragmatic intercostal hernia of bowel loops and omentum with herniation of lung—a very rare entity . Egypt J Chest Dis Tuberc 2016 ; 65 : 545 – 8 . Google Scholar Crossref Search ADS 10 Lasithiotakis K , Venianaki M , Tsavalas N , Zacharioudakis G , Petrakis I , Daskalogiannaki M et al. Incarcerated spontaneous transdiaphragmatic intercostal hernia . Int J Surg Case Rep 2011 ; 2 : 212 – 14 . Google Scholar Crossref Search ADS PubMed 11 Cole FH Jr , Miller MP , Jones CV. Transdiaphragmatic intercostal hernia . Ann Thorac Surg 1986 ; 41 : 565 – 6 . Google Scholar Crossref Search ADS PubMed 12 Forty J , Wells FC. Traumatic intercostal pulmonary hernia . Ann Thorac Surg 1990 ; 49 : 670 – 1 . Google Scholar Crossref Search ADS PubMed 13 Henriques AC , Malena CR , Freitas AC , Waisberg J , Pires AC . Transdiaphragmatic intercostal hernia after spontaneous rib fractures secondary to coughing fit . Rev Col Bras Cir 2010 ; 37 : 78 – 80 . Google Scholar Crossref Search ADS PubMed 14 Kalles V , Dasiou M , Doga G , Papapanagiotou I , Konstantinou EA , Mekras A et al. Posttraumatic transdiaphragmatic intercostal hernia: report of a case and review of the literature . Int Surg 2015 ; 100 : 444 – 9 . Google Scholar Crossref Search ADS PubMed 15 Kao P , Fang H-Y , Lu T-Y , Hsu S-C , Chen C-K , Chen P-R. Strangulation of chronic transdiaphragmatic intercostal hernia . Ann Thorac Surg 2014 ; 97 : e155 – 7 . Google Scholar Crossref Search ADS PubMed 16 Sharma OP , Duffy B. Transdiaphragmatic intercostal hernia: review of the world literature and presentation of a case . J Trauma 2001 ; 50 : 1140 – 3 . Google Scholar Crossref Search ADS PubMed 17 Hruska LA , Corry D , Kealey GP. Transdiaphragmatic intercostal hernia resulting from blunt trauma: case report . J Trauma 1998 ; 45 : 822 – 4 . Google Scholar Crossref Search ADS PubMed 18 Saw EC , Yokoyama T , Lee BC , Sargent EN. Intercostal pulmonary hernia . Arch Surg 1976 ; 111 : 548 – 51 . Google Scholar Crossref Search ADS PubMed 19 Biswas S , Keddington J. Soft right chest wall swelling simulating lipoma following motor vehicle accident: transdiaphragmatic intercostal hernia. A case report and review of the literature . Hernia 2008 ; 12 : 539 – 43 . Google Scholar Crossref Search ADS PubMed 20 Anderson RH , Shirali G. Sequential segmental analysis . Ann Pediatr Card 2009 ; 2 : 24 – 35 . Google Scholar Crossref Search ADS 21 Meinberg E , Agel J , Roberts C , Karam MD , Kellam JF. Fracture and dislocation classification compendium . J Orthop Trauma 2018 ; 32(1 Suppl) : S161 – 5 . 22 Moore EE , Malangoni MA , Cogbill TH , Shackford SR , Champion HR , Jurkovich GJ. Organ injury scaling IV: thoracic vascular, lung, cardiac and diaphragm . J Trauma 1994 ; 36 : 229 – 300 . 23 Moore EE , Cogbill TH , Jurkovich GJ , McAninch JW , Champion HR , Gennarelli TA et al. Organ injury scaling III: chest wall, abdominal vascular, ureter, bladder and urethra . J Trauma 1992 ; 33 : 337 – 9 . Google Scholar Crossref Search ADS PubMed 24 Schulz-Drost S , Syed J , Besendoerfer M , Mauerer A , Blanke M , Schulz-Drost M et al. Elastic stable chest wall repair as a means of stabilizing the anterior chest wall in recurrent pectus excavatum with sternacostal pseudarthrosis: an innovative fixation device . Thorac Cardiovasc Surg 2015 ; 63 : 419 – 26 . Google Scholar PubMed 25 Sollender GE , White TW , Pieracci FM. Fracture of the costal cartilage: presentation, diagnosis, and management . Ann Thorac Surg 2018 ; pii: S0003-4975(18)31499-1. doi:10.1016/j.athorasur.2018.08.076. © The Author(s) 2019. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Unifying classification for transdiaphragmatic intercostal hernia and other costal margin injuries
JF - European Journal of Cardio-Thoracic Surgery
DO - 10.1093/ejcts/ezz020
DA - 2019-07-01
UR - https://www.deepdyve.com/lp/oxford-university-press/unifying-classification-for-transdiaphragmatic-intercostal-hernia-and-oHjoc3V0z8
SP - 150
VL - 56
IS - 1
DP - DeepDyve
ER -