Peri-operative course of peritonitis following tube thoracostomy: A misdiagnosed case of congenital diaphragmatic hernia.

A young adult presented with signs of peritonitis following tube thoracostomy for suspected acute hydropneumothorax. Attempted decompression of the chest by tube thoracostomy had caused gastric perforation, and on surgical exploration, he was found to have a congenital diaphragmatic hernia with herniation of the stomach, spleen and colon. All intensive care doctors and emergency room physicians dealing with the care of patients with acute respiratory failure should be taught to recognize and keep the possibility of a Bochdalek hernia in mind, especially in young adults presenting with unusual respiratory and gastrointestinal symptoms.

INTRODUCTION

Diaphragmatic hernias are rarely seen in adults. Of those that occur, the majority are hernias secondary to acute traumatic disruptions of the diaphragm and hiatus hernias. A small number of diaphragmatic hernias in adults represent congenital hernias, namely Morgagni or Bochdalek hernias. Though most congenital diaphragmatic hernias (CDH) are symptomatic at birth and receive surgical correction, a few remain silent and these are usually detected incidentally in adulthood. Rarely, they become symptomatic and present in several unusual ways posing a diagnostic dilemma to the attending physician.

CASE REPORT

A 22-year-old male presented to a primary care hospital with pain and tightness in the left upper abdomen along with shortness of breath of two days duration. He had past history of repeated episodes of mild to moderate breathlessness for which he consulted local physicians and was prescribed symptomatic treatment. He belonged to a lower income family and was a daily wage earner because of which he did not get himself evaluated in greater detail in the past. During the current presentation, a chest radiograph [Figure 1] revealed an air fluid level in the left hemithorax. This was diagnosed as hydropneumothorax and a left-sided intercostal drain (ICD) was inserted which drained serosanguinous fluid. As the intercostal drains had not relieved his symptoms, a second ICD was inserted which began to drain fluid stained with blood. There was temporary relief of respiratory symptoms for a day; however, low-grade fever with pleuritic chest pain followed and he was referred to our hospital. During the two days before admission to our hospital, he gave the history that he was unable to pass stool or flatus and had several episodes of non-bilious vomiting. There was no referral note with the patient, hence further details of clinical findings and treatment were not available.

Figure 1

Air-fluid level in the left hemithorax with mediastinal shift towards the right side resembling left hydropneumothorax. On careful inspection a curvilinear line is noted (white arrows) which represents the gastric outline with lung markings above it

On examination, he was febrile but hemodynamically stable. There were two left ICDs in the left hemithorax draining turbid, foul-smelling contents. Air entry was decreased on the left side of the lung, which was radiologically found to be due to the presence of abdominal viscera and fluid in the left hemithorax. There was abdominal tenderness with guarding and rigidity and moderate respiratory embarrassment in the form of using of intercostal muscles of respiration along with subphrenic recession. Bowel sounds were absent. His pulse rate was 122 per min and blood pressure was 90 mm Hg systolic, which improved to 110/80 mm Hg after fluid resuscitation in the emergency room. His respiratory rate was 34–36 per min and pulse oximetry saturation was 92–94%.

Blood investigations were unremarkable. Blood gas analysis showed mild hypoxemia (paO2 72 mm Hg on a fractional inspiratory oxygen concentration of 35%) and mild metabolic acidosis, pH 7.31 with serum bicarbonate of 18 mmoles/liter. A chest radiograph revealed an air-fluid level in the left hemithorax with two ICDs in situ [Figure 2]. The right lung field was normal.

Figure 2

Following tube thoracostomy chest tubes are seen within the gut lumen (black arrows). The gastric outline is marked with the white arrow

Since the patient had signs of peritonitis and sepsis, he was taken for an emergency exploratory laparotomy. General anesthesia was induced and crash induction was performed with injection ketamine 2 mg/kg, injection fentanyl and injection succinyl choline intravenously. He was intubated with a cuffed oral endotracheal tube, size 8.5 and maintained on oxygen, nitrous oxide and isoflurane along with muscle relaxant vecuronium bromide. Intravenous fluids, lactated ringers and dextrose were infused. At surgery, a 20×10 sq cm defect was noted in the posterolateral part of the left dome of the diaphragm. The entire stomach, the splenic flexure of the colon and the spleen had herniated into the left side of the chest. There was no peritoneal sac over the stomach which signed the congenital nature of the hernia. The two ICDs were found to have traumatized the spleen and stomach and were draining blood and gastric contents. The tubes were disconnected and removed. A splenectomy was done, the hernial contents were reduced and the perforation in the stomach was closed. An ICD was placed in the pleural cavity and the diaphragmatic defect was repaired by primary closure. He required transfusion of two units of packed red blood cells during the surgery. The patient was shifted to the intensive care unit post operatively for respiratory and hemodynamic support without reversal of neuromuscular blockade. Immediate postoperative chest radiograph revealed mild pleural effusion in the left hemithorax. Over the next three to five days, both lobes of the lung were found to have re-expanded on serial chest radiographs. There were no findings suggestive of lung hypoplasia. He was subsequently extubated on the fifth postoperative day and remained hemodynamically stable during the postoperative course. Chest expansion techniques and physiotherapy helped to aid in his recovery. Subsequently he was transferred to the general ward on the seventh postoperative day and he recovered without further complications.

DISCUSSION

Mistaking a diaphragmatic hernia for a hydropneumothorax in a patient presenting with acute respiratory distress though rare, is not an isolated occurrence. In adults, the diagnosis of CDH is usually missed unless there is a high index of suspicion. Thomas et al., have found that nearly 38% of adults with CDH are misdiagnosed in this way.[1] In adults, the reported incidence of CDH is varied, with estimates ranging from 1 in 2000-7000 based on autopsy findings to 0.17-6% based on computerized tomography.[2] Delayed presentations with chronic respiratory or gastrointestinal complaints have been reported from one month to late adulthood in 5–25% of patients.[3] In the event of a sudden herniation, the patient may experience severe respiratory distress or symptoms of acute intestinal obstruction. In such cases, the massively distended stomach and abdominal viscera within the hemithorax may easily be mistaken as pneumothorax, pleural effusion or hydropneumothorax as in our patient.

Elhalaby et al., studied 33 adult patients with CDH between 1993-2000, 10 of whom had chronic respiratory symptoms or gastrointestinal obstruction. Inappropriate tube thoracostomy was done in three of those patients (i.e. 30% of patients who presented), two of who were diagnosed as having pleural effusion and one as pneumothorax.[4] Rege et al., described a 26-year-old woman who presented with chest pain following minor trauma. The chest radiograph showed gastrothorax which (as in this case) was misinterpreted as hydropneumothorax and the stomach was perforated by tube thoracostomy.[5]

A CDH should be suspected in patients presenting with respiratory symptoms, especially if the gastric air bubble cannot be seen in the ′regular′ position on a chest radiograph. Closer examination of the chest radiograph in our patient [Figure 1] showed lung markings in the left upper zone, separated from the large lucent area below by a curved radio-opaque line representing the stomach wall. These changes are not consistent with a diagnosis of hydropneumothorax, in which the peripheries of the lung fields will be devoid of lung markings and the collapsed lung margin will be visible. Additional lateral decubitus and true erect frontal chest radiographs may aid in the diagnosis as there will not be migration of the “fluid level” with change in posture. Insertion of a nasogastric tube may be useful as it will be seen within the “gas bubble” in the thorax. A “pleural fluid” analysis to ascertain the nature of the thoracic contents may be a simple and useful test in doubtful cases before a definitive procedure such as a tube thoracostomy is carried out. If gastric contents are suspected, one should proceed to investigate further with other imaging techniques.

In conclusion, the present case highlights the alternative possible diagnosis that an intensivist or an emergency care physician needs to keep in mind in a patient presenting with acute respiratory failure and chest radiograph suggestive of an apparent air-fluid level. Here, the CDH with its contents may easily be mistaken for a pneumothorax, pleural effusion or hydropneumothorax unless a careful and detailed examination is made as well a high index of suspicion is carried. A wrong and hasty diagnosis, such as what happened in this case, could lead to wrong treatment being applied resulting in prolonged morbidity and a complicated and expensive course. Hence, one should keep in mind the incidence of silent adult congenital diaphragmatic hernias when a young patient presents with unusual respiratory symptoms, even without a history of trauma. In addition, the presence of a pleural effusion or hydropneumothorax on chest radiograph should not immediately prompt the placement of a chest tube, but a careful and thorough examination of the patient before any course of action. The key issue is prompt identification of the condition and rapid surgical correction.