Current value of peritoneal tap in blunt abdominal trauma

EUROPEAN JOURNAL OF EMERGENCY MEDICINE, 2002, 9, 253^257 Current value of peritoneal tap in blunt abdominal trauma Ü. ERGENE1*, F. COS,KUN1, O. ERAY1, Ö GÖKÇE2, J. FOWLER1, M. HACIYANLI2, Z. TAS,AR1 and N. NUR USER1 1 Department of Emergency Medicine, and Balçova, Izmir, 35140 Turkey 2 Department of Surgery, Dokuz Eylül University School of Medicine, This study aimed to establish the diagnostic value of paracentesis (peritoneal tap) in the assessment of patients with blunt abdominal trauma. Paracentesis, using a four-quadrant puncture technique, was performed in blunt abdominal trauma victims presenting to the emergency department of a tertiary-care university medical centre. Pregnant patients, those under 18 or those having an abdominal scar were excluded from the study. All patients then underwent one of the following procedures as indicated: emergency ultrasound, abdominal computed tomography scan, diagnostic peritoneal lavage or laparotomy. Paracentesis results were compared with the results of other tests and surgery in diagnosing haemoperitoneum. Haemoperitoneum was confirmed surgically in six of the seven patients with a positive paracentesis. Nine out of 65 patients with positive clinical findings but negative taps underwent surgical intervention, and abdominal bleeding was confirmed in eight. Three seriously injured patients died before diagnostic studies or laparotomy could be performed. In conclusion, a positive paracentesis result may be used to guide decision-making in the setting of blunt abdominal trauma if other diagnostic methods are unavailable. Its high false-negative rate limits its overall usefulness. & 2002 Lippincott Williams & Wilkins. Keywords: blunt abdominal trauma; paracentesis; haemoperitoneum INTRODUCTION Blunt abdominal trauma patients account for a substantial portion of those emergency patients that have a poor prognosis. Accurate diagnosis and early surgical intervention in selected cases improve the chances for a good outcome. Using recently developed diagnostic tools, the early and accurate diagnosis of blunt abdominal trauma is relatively easy in advanced trauma centres. However, in many countries around the world, paracentesis (peritoneal tap) is widely utilized, especially after normal working hours when access to diagnostic imaging modalities may be limited. Few studies regarding paracentesis in trauma patients have been published during the last 35 years. In fact, paracentesis for trauma patients is only mentioned as a matter of historic interest in current textbooks.1 However, * To whom correspondence should be addressed at Seferihisar Caddesi, Dogakent Sitesi No. 8, GüzelbahçeIzmir, Turkey & 2002 Lippincott Williams & Wilkins. because of its continued widespread utilization, we decided to study the accuracy of this method compared with other modern diagnostic tools. Abdominal paracentesis was introduced by Salomon in 1906,2 and was then refined by many others. Because of its high false-negative rate in the diagnosis of haemoperitoneum, the technique was largely abandoned.3,4 When 200 ml of fluid is present in the peritoneum, only 20% of taps are positive, and when 500 ml is present, 80% are positive.3 With the onset of the quantitative technique of peritoneal lavage, the high false-negative rate associated with paracentesis spelled its demise in modern emergency medicine practice. Some general surgeons still prefer paracentesis because of its simplicity and time-saving features compared with diagnostic peritoneal lavage. With its simple instrumentation using a 10 ml syringe and a 20 gauge needle, its relative ease in severely ill patients necessitating bedside application, and low complication rate, the technique is still gaining interest. 254 ERGENE, COS , KUN, ERAY, GKCE, FOWLER, HACIYANLI,TAS , AR and NUR USER In this study, we aimed to measure the accuracy of paracentesis for diagnosing traumatic haemoperitoneum in the emergency department setting. MATERIALS AND METHODS Study population Blunt abdominal trauma patients (61 male and 11 female) presenting to the emergency department while one of the primary researchers was on duty were considered for inclusion in the study. Those patients who were under the age of 18 years, pregnant or who had a previous laparotomy scar were excluded from the study. Study design Initial resuscitation was performed by the emergency medicine resident, and then orthopaedic, surgery, anaesthesiology or neurosurgery consultations were obtained when indicated. Paracentesis was performed by the emergency medicine resident in all the study patients. If the patient was haemodynamically stable, diagnostic ultrasound (US) or computed tomography (CT) scan was performed and read by a radiologist. Haemodynamically unstable patients underwent diagnostic peritoneal lavage (DPL) performed by general surgery residents. Patients with a positive DPL or positive CT or/and US underwent surgery, and patients with a negative DPL or negative CT or/and US were observed for 8–48 h depending on the clinical symptoms. Findings at laparotomy were compared with the results of the paracentesis and other diagnostic tests in patients undergoing surgery. Procedures Paracentesis was performed as customarily practised in Turkey. With the patient in the supine position, the skin was cleansed with povidine–iodine solution and the peritoneum entered along the lateral margin of the rectus muscle using a 20 gauge needle. If no blood return was obtained after gentle aspiration, the needle tip was repositioned at a new angle and the aspiration was repeated. This procedure was repeated in each abdominal quadrant. The tap was considered positive if an abnormal substance (bile, faecal material, urine) or more than 0.5 ml of non-clotting blood was aspirated. DPL was performed in patients with an unstable haemodynamic status. In keeping with standard percutaneous DPL recommendations5 and using sterile techniques, a 15 cm disposable catheter was inserted into the peritoneum under local anaesthesia through an infraumbilical incision. A positive DPL EUROPEAN JOURNAL OF EMERGENCY MEDICINE (2002) 9(3) was defined as the aspiration of abnormal fluid. If less than 10 ml of free blood was withdrawn through the catheter, 1 l of normal saline was instilled into the peritoneal cavity and the effluent sent for red blood cell count and determination of amylase and alkaline phosphatase levels. The macroscopic presence of blood, bile, ingested food particles or faecal material in the effluent was accepted as a positive DPL result.6,7 Data analysis Sensitivity, specificity, and positive and negative predictive values were assessed. RESULTS Paracentesis was positive in seven of the 72 patients. Six of these patients underwent laparotomy, and abdominal bleeding was confirmed in all of these (Table 1). In one patient with a positive peritoneal tap, other diagnostic tests did not support the diagnosis of haemoperitoneum, and he improved with supportive therapy and was discharged without complications. Of the seven paracentesis-positive patients, three were haemodynamically stable and four were unstable (Table 2). Among the 65 patients with a negative paracentesis, nine were operated on because of strong clinical suspicion and positive results of diagnostic tests (US, CT or DPL). Haemoperitoneum was confirmed by laparotomy in eight of these nine patients (Table 3). Three of the paracentesis-negative patients undergoing laparotomy were unconscious (Table 4). In summary, six out of seven patients with a positive paracentesis, and eight out of 65 with a negative paracentesis had haemoperitoneum (Table 5). Table 1. Diagnostic test results in paracentesis-positive patients Surgical diagnosis Paracentesis US CT DPL Major vessel rupture Liver, spleen, IVC and right kidney lacerations Liver laceration Liver haematoma Spleen laceration Liver laceration No intra-abdominal abnormality (abdominal findings resolved without any intervention) þ þ þ þ þ þ þ þ þ þ þ þ þ   þ þ  þ and  indicate abnormal and normal results, respectively. IVC, inferior vena cava; US, abdominal ultrasound; CT, computed tomography; DPL, diagnostic peritoneal lavage. PERITONEALTAP IN BLUNTABDOMINALTRAUMA 255 Table 2. Surgical findings and other clinical information in paracentesis-positive patients Surgical diagnosis Vital signs LOC Other injuries Major vessel rupture Liver, spleen, IVC and right kidney lacerations Liver laceration Liver haematoma Unobtainable BP Systolic BP 40 mmHg Coma Awake Stable Stable Awake Awake Spleen laceration Liver laceration No intra-abdominal abnormality (abdominal findings resolved without any intervention) Stable BP 80/60 mmHg Stable Awake Awake Coma Pelvis Head, neck, thorax, extremities, vertebra Head, neck, thorax Thorax, vertebra (operation performed on third day after haemoglobin dropped) None Fracture of extremities Head, thorax IVC, inferior vena cava; BP, blood pressure; LOC, level of consciousness. Table 3. Diagnostic test results in paracentesis-negative patients undergoing laparotomy Definitive diagnosis Paracentesis US CT DPL Contusion of the spleen Liver and spleen lacerations Splenic rupture Splenic rupture Spleen and pancreas laceration; splenic vein rupture Mesenteric artery rupture Splenic rupture Rupture of the spleen, liver and diaphragm No abnormality on laparotomy      þ     þ  þ þ þ þ þ  Abdominal US was performed in four of the seven patients with positive taps, and revealed peritoneal fluid in three. The patient with a negative US also had a negative CT and a negative DPL. Abdominal US was performed in 55 patients with negative taps, and was positive in four patients. All of these patients underwent laparotomy and haemoperitoneum was found. Of the 51 patients with negative paracentesis, there was one with negative US; abdominal bleeding was diagnosed with DPL and confirmed by surgery. þ þ þ þ and  indicate abnormal and normal results, respectively. US, abdominal ultrasound; CT, computed tomography; DPL, diagnostic peritoneal lavage. CT was performed in 11 patients with negative tap results and was positive for peritoneal fluid in one patient. This patient had haemoperitoneum at surgery. In one patient, CT and US were negative, whereas the DPL was positive. This patient had also haemoperitoneum at surgery. The remaining nine patients with normal CT scans were followed clinically without complications. Of the seven paracentesis-positive patients, five underwent DPL. In one of these patients, the DPL was negative, and the patient was followed without surgical intervention and developed no complications. The other four DPL-positive patients were found to have haemoperitoneum on laparotomy. Seventeen patients with negative taps underwent DPL; five of these were positive. Surgery confirmed the bleeding in four DPL-positive patients and one had no haemoperitoneum. Three of our patients died without having an autopsy or laparotomy. One had a positive paracentesis but was followed without laparotomy because of a stable haemodynamic condition and other negative diagnostic test results. However, the patient died from major head trauma. A second patient with a positive paracentesis underwent DPL, which revealed gross haemorrhage, but the patient died as she was taken emergently to the operating theatre. A third patient had a negative paracentesis and a positive DPL, but the patient died before surgery could be performed. These three patients were excluded from our study results because their families did not give permission for an autopsy to be performed. Abdominal CT was positive in one of the two patients with positive tap results. Haemoperitoneum was then surgically confirmed in this patient. The CT-negative but paracentesis-positive patient also had the negative DPL result, and was followed without surgery. Of the 65 patients with negative paracentesis results, 23 had negative results on other diagnostic tests. These patients remained clinically stable after several hours of observation and were discharged home. No complications were reported on follow-up. Another Overall, the sensitivity, specificity, and positive and negative predictive values of paracentesis were found to be 0.43, 0.98, 0.86 and 0.88, respectively. EUROPEAN JOURNAL OF EMERGENCY MEDICINE (2002) 9(3) ERGENE, COS , KUN, ERAY, GKCE, FOWLER, HACIYANLI,TAS , AR and NUR USER 256 Table 4. Definitive diagnosis and other clinical information in paracentesis-negative patients undergoing laparotomy Definitive diagnosis Vital signs LOC Other injuries Contusion of the spleen Liver and spleen lacerations Splenic rupture Splenic rupture Spleen and pancreas laceration; splenic vein rupture Mesenteric artery rupture Splenic rupture Rupture of the spleen, liver and diaphragm No abnormality on laparotomy Stable Stable Stable Stable Stable Stable Stable Stable Stable Awake Awake Coma Awake Awake Lethargy Awake Awake Coma Head, thorax Thorax, extremities, vertebra Thorax Head, vertebra Head Head, thorax, extremities Head, extremities Thorax Head, thorax LOC, level of consciousness. Table 5. Peritoneal findings versus paracentesis results Paracentesis þ Paracentesis – Haemoperitoneum þ Haemoperitoneum – Total 6 8 1 57 7 65 33 patients were admitted to the hospital for management of non-abdominal traumatic lesions (trauma to head, thorax, extremities, vertebral or pelvis). No complications of paracentesis were observed in these patients. DISCUSSION When paracentesis was introduced by Salomon in 1906, it was proclaimed revolutionary. Accuracy rates of up to 75–88% for the detection of haemoperitoneum have been reported.7–10 Our study confirms the high (0.86) positive predictive value of paracentesis in blunt trauma victims. Because haemoperitoneum does not necessarily reflect continued intra-abdominal bleeding, taking the patient’s clinical status into consideration is important when making clinical management decisions. It was found that 200 ml of fluid in the abdominal cavity resulted in a sensitivity of only 20% for paracentesis, whereas the sensitivity with 500 ml of abdominal fluid was 78%. Because of its low sensitivity, and the development of the more sensitive DPL technique in the early 1960s, paracentesis was abandoned in most modern medical centres.3,4 With the refinement of non-invasive imaging studies such as US and CT, DPL is also being used less and less in diagnostic strategies in blunt trauma patients. The presence of abdominal bleeding in eight out of 65 patients with a negative paracentesis reveals the major weakness of the technique, that is its low sensitivity. When incorporated into a diagnostic strategy for blunt trauma patients, a negative paracentesis has to be verified by other diagnostic procedures. One group found that the sensitivity of EUROPEAN JOURNAL OF EMERGENCY MEDICINE (2002) 9(3) DPL at detecting intra-abdominal injury was higher than the paracentesis group (98% versus 72%) in 566 trauma patients.4 In another study, the overall predictive value of the technique was found to be 80%.11 Paracentesis has been reported to have a low complication rate. No complications occurred amongst our 72 patients. Other researchers also have reported on the safety of the technique.8,12 Most researchers have used a two- or four-quadrant puncture technique. We used the four-quadrant technique, as this technique is widely preferred in our country.13 Because of its simplicity, lack of complications and relatively high specificity (98%), paracentesis continues to be used in our country, where US and CT are often not readily available. It should not be performed or relied on in centres that have more sensitive diagnostic imaging techniques (e.g. US, CT) available. Although all of the sensitive imaging techniques such as US, CT and magnetic resonance imaging (MRI) have been available in our hospitals, many surgeons used paracentesis routinely in this setting during the study period. For this reason we decided to compare the results of paracentesis with those of other diagnostic techniques and with the results of laparotomy. REFERENCES 1. Knudson, M.M. and Maull, K.I. (1999) Nonoperative management of solid organ injuries. Surg. Clin. N. Am., 79(6), 1357–71. PERITONEALTAP IN BLUNTABDOMINALTRAUMA 2. Salomon, H. (1906) Die Diagnostische Punktion des Bauches. Berl. Klin. Wochenschr., 43, 43–5. 3. Hedges, R. (1991) Paracentesis. In: Clinical procedures in emergency medicine. Philadelphia: WB Saunders, pp. 674–9. 4. Nagy, K.K., Fildes, J.J., Sloan, E.P., Kim, D.O., Smith, R.F., Robert, R.R., et al. (1995) Aspiration of free blood from the peritoneal cavity does not mandate immediate laparotomy. Am. Surg., 9, 790–5. 5. Burch, J.M., Franciose, R.J. and Moore, E.E. (1999) Trauma. In: Schwartz, S.I., Shire, G.T., Spencer, F.C., Daly, J.M., Fisher, J.E., Galloway, A.C., editors. Principles of surgery. New York: McGraw-Hill, pp. 155–6. 6. McAnena, O.J., Marx, J.A. and Moore, E.E. (1991) Peritoneal lavage enzyme determinations following blunt and penetrating abdominal trauma. J. Trauma., 31, 1161–4. 7. Otomo, Y., Mashiko, K., Morimura, N. and Otsuka, T. (1989) Usefulness and problems of peritoneal tap and lavage on the diagnosis of blunt abdominal trauma – 257 8. 9. 10. 11. 12. 13. efficacy for diagnosis of intestinal injury. Nippon Geka Gakkai Zasshi., 90, 2008–14. Mahanta, H., Das, M.K. and Dutta Choudhury, S.B. (1990) An experience with diagnostic paracentesis in 100 cases of acute abdomen. J. Indian. Med. Assoc., 88, 125–8. Davis, J.J., Cohn, I. and Nance, F.C. (1975) Diagnosis and management of blunt abdominal trauma. Ann. Surg., 183, 672–8. Yurko, A.A. and Williams, R.D. (1996) Needle paracentesis in blunt abdominal trauma: a critical analysis. J. Trauma., 6, 194–7. Drapanas, T. and McDonald, J. (1961) Peritoneal tap in abdominal trauma. Surgery., 50, 742. Gupta, S., Talwar, S., Sharma, R.K., Gupta, P., Goyal, A. and Prasad, P. (1996) Blunt trauma of the abdomen: a study of 63 cases. Indian J. Med. Sci., 50, 272–6. Mansoor, T., Zubari, S. and Masiullah, (2000) Evaluation of peritoneal lavage and abdominal paracentesis in cases of blunt abdominal trauma – a study of fifty cases. J. Indian Med. Assoc., 98, 174–5. EUROPEAN JOURNAL OF EMERGENCY MEDICINE (2002) 9(3)