Article Text
PDF
Abstract
Objectives The civil war in Syria began on 15 March 2011, and many of the injured were treated in the neighbouring country of Turkey. This study reports the surgical outcomes of this war, in a tertiary centre in Turkey.
Methods 159 patients with civilian war injuries in Syria who were admitted to the General Surgery Department in the Research and Training Hospital of the Medical School of Mustafa Kemal University, Hatay, Turkey, between 2011 and 2012 were analysed regarding the age, sex, injury type, history of previous surgery for the injury, types of abdominal injuries (solid or luminal organ), the status of isolated abdominal injuries or multiple injuries, mortality, length of hospital stay and injury severity scoring.
Results The median age of the patients was 30.05 (18–66 years) years. Most of the injuries were gunshot wounds (99 of 116 patients, 85.3%). Primary and previously operated patients were transferred to our clinic in a median time of 6.28±4.44 h and 58.11±44.08 h, respectively. Most of the patients had intestinal injuries; although a limited number of patients with colorectal injuries were treated with primary repair, stoma was the major surgical option due to the gross peritoneal contamination secondary to prolonged transport time. Two women and 21 men died. The major cause of death was multiorgan failure secondary to sepsis (18 patients).
Conclusions In the case of civil war in the bordering countries, it is recommended that precautions are taken, such as transformation of nearby civilian hospitals into military ones and employment of experienced trauma surgeons in these hospitals to provide effective medical care. Damage control procedures can avoid fatalities especially before the lethal triad of physiological demise occurs. Rapid transport of the wounded to the nearest medical centre is the key point in countries neighbouring a civil war.
- TRAUMA MANAGEMENT
Statistics from Altmetric.com
Key messages
The civil war in a neighbouring country causes complex problems especially in medical support.
Damage control surgery is important to avoid poor outcomes.
Transfer time of an injured patient to a medical centre in a neighbouring country may be quite prolonged.
Transformation of nearby civilian hospitals to military ones to decrease transfer time and to perform adequate interventions may decrease mortality and morbidity.
Introduction
Wars are major tragedies for all participants and have negative effects on the neighbouring countries as immigration due to war often causes social and economic problems; conversely, medical support by a neighbouring country may also affect its own medical and economic capacity.1 The civil war in Syria began on 15 March 2011, and although there are no exact data about casualty rates, >191 000 civilians and soldiers have died and many more have been injured by gunshot and bombs in Syria according to the Office of the UN High Commissioner for Human Rights.2 Although many of the injured patients were treated in Syria, some were transferred to hospitals in neighbouring countries. Turkey is one of the nearest countries to the battlefield in Syria (Figure 1),3 and surgeons in Turkey, especially in nearby cities such as Hatay, Kilis, Gaziantep and Urfa, have been performing significant numbers of operations on civil war victims.
The map of the Turkey–Syria borderline.
Published studies about war surgery tend to report the experiences around the battlefield.4–6 The mortality rate is higher in these studies since they analyse the data of injured patients in the country of war and the casualties occurred within the first 30 min. This study uniquely reports the outcomes of civil war casualties treated in a neighbouring country.
Materials and methods
All patients who sustained injuries due to the civil war in Syria who were admitted to the Department of General Surgery in the Research and Training Hospital of the Medical School of Mustafa Kemal University, Hatay, Turkey, between March 2011 and December 2012 were studied retrospectively. Some of these patients had already been operated on in Syria—those who did not need any further surgical intervention after arrival in our hospital were subsequently excluded from the analysis. The medical notes were examined to identify the sex, age, injury type, history of previous surgery for the injury, abdominal injuries (solid vs hollow organ), the status of isolated abdominal injuries or multiple injuries, mortality, length of hospital stay and injury severity scoring. Trauma scores were evaluated according to the Abbreviated Injury Scale and Injury Severity Score ().
Statistical analysis
Statistical analyses were performed using the SPSS software V.15. The variables were investigated using visual (histograms, probability plots) and analytic (Kolmogorov–Smirnov/Shapiro–Wilk's test) tests to determine whether or not they were normally distributed. Descriptive analyses were presented using medians and IQR for the non-normally distributed and ordinal variables. Since the hospital stay times were not normally distributed, non-parametric tests were conducted to compare these parameters, as well as to compare ordinal variables. The Mann–Whitney U test was used to compare hospital stay time and admission type between the groups. ISSs of the patients were not distributed normally, hence Mann–Whitney U test was used to compare ISS, mortality and injury type between groups. Since transfer time values were normally distributed, the Student's t test was performed to compare these parameters between different groups according to mortality and hospital stay time. Patients' admission type and injured organs were compared between different groups according to mortality using the χ2 test. A p value <0.05 was considered as statistically significant.
Results
In the 20 months of the study, 159 patients with Syrian civil war injuries were admitted to the general surgical department. Seventy-nine patients had been primarily operated in Syria, and 43 of these did not require any additional surgery in Turkey and were excluded from further analysis, leaving 116 patients with gunshots or fragment injuries in the study—80 received their first surgery in Turkey and 36 had previously been operated on in Syria but underwent further surgery in Turkey. The majority of casualties were men (109/116 93.9%), the median age was 30 (18–66 years) years and 99/116 (85.3%) were gunshot injuries—the remainder being from bombs.
The transfer time of primarily transported patients to our hospital from the battlefield was 6.28±4.44 h. On the other hand, the previously operated patients were transferred to our department in a median time of 58.11±44.08 h after their first surgical interventions. The median hospital stay time was 8.0 (3–95) days and 16.5 (3–36) days for primarily operated patients and patients who received an initial intervention and transferred, respectively, and there was a statistically significant difference between these two groups regarding the hospital stay time (p<0.05).
As there were no recorded medical data regarding previous surgery in Syria, all of the transferred 36 patients after initial assessments received re-laparotomies according to their clinical situation in our clinic, and 6 of them were negative, 12 had undisclosed abdominal packing and the remaining 18 patients had hollow visceral injury or anastomotic leaks with gross intra-abdominal contamination.
Fifty-four (46.6%) patients had only one abdominal organ injury (solid or hollow), and the rest had multiple organ injuries in various combinations (Table 1), whereas 64 patients had isolated abdominal injuries and 52 patients had multisystem trauma. The most coexisting injuries were chest and extremity traumas (Table 2). There were two amputations in the lower extremities and one amputation of both hands due to fragment injury, all of which were performed in Syria. The rest of the orthopaedic injuries were bony fractures from gunshots or fragment injuries. The most common chest traumas were reported after fragment injury, and haemopneumothorax was the leading lesion. Two patients had undergone lobectomy and four had diaphragmatic injuries repaired with primary sutures. There were no statistically significant differences in mortality between isolated or multiple trauma patients.
The number of injured organs (some patients had injuries to more than one viscera)
Other injuries accompanying abdominal traumas
The most frequently performed surgical procedures were stomas. Twenty-four patients received ileostomy/jejunostomy and 23 patients received colostomy as well as various anastomoses and/or other surgical interventions (Table 3). There were 10 rectal injuries during this period and 9 were treated by a Hartmann's procedure. Primary repair of rectum was performed in one injured patient with a small perforation due to a fragment injury and without severe intra-abdominal contamination. Colon injuries were mostly treated by stomas—only 13/44 colon perforations were managed without any stomas. There were four anastomotic leaks out of nine colon injuries treated by resection and anastomosis and a colocutaneous fistula in one of the four primary colonic repairs, and all of these five patients received stomas after re-laparotomies. Double barrel ostomies or mucus fistula were the preferred options during re-laparotomies for anastomotic leaks or colocutaneous fistula. Overall, 6/56 small bowel injuries were treated with diversion stomas after primary repair or anastomosis since they were located in the terminal ileum and gross peritoneal contamination was present; 6 of the 50 small bowel injuries initially treated without a stoma subsequently leaked and required re-laparotomy and diversion jejunostomy. Septic complications occurred in eight patients with failed intestinal surgeries, and five of them died due to multiorgan failure secondary to sepsis.
Surgical procedures performed in our clinic on Syrian civil war casualties during 2011–2012
The overall median ISS was 18.00 (9.00–61.00) with fragment and gunshot wound median ISSs of 20 (9–54) and 16 (9–61), respectively, but this difference was not significant. Unsurprisingly there was a significant difference (p<0.05) between the median ISS of those who died and survived to discharge (34 vs 16).
Two of seven women and 21/109 men died. Also, 12 of the 36 cases (33.3%) who were transferred after surgery in Syria died compared with 11/80 (13.7%) for those patients who were primarily treated in our clinic, which was not significant. The major cause of death was multiorgan failure secondary to sepsis in 18 patients; only three of the patients died due to haemorrhagic shock.
Discussion
Military surgery has many differences to civilian trauma surgery. Weapons of war cause more serious high kinetic energy injuries, mines or bombs lead to more destructive effects via high-energy transfer, the transfer time of injured cases to medical care can be prolonged and the incidence of infection is higher.6 Ideally, interventions to the wounded patients during war should be performed by military surgeons, but in reality civil surgeons are often obliged to treat such patients nearly all around the world.7 Turkey is one of the neighbouring countries to Syria, and the surgeons in the border cities continue to treat many of the wounded cases from the ongoing civil war.
There are many hospitals in Hatay, and our hospital, the Research and Training Hospital of Medical School of Mustafa Kemal University, situated approximately 70 km from the Syrian border, is one of the most developed medical centres of the region and is a Level I Trauma Centre with 500 beds, having nearly all medical disciplines. Before the Syrian civil war, only a small number of trauma patients were being treated in our department (15–30 cases/year) and almost none of the surgeons had high-intensity trauma experience.
The colon is the most commonly injured organ in penetrating abdominal trauma in both civilian and war trauma,6 ,8 although small bowel injuries were commonest in this study with colonic trauma second. Diversion stomas for colorectal injuries can cause psychological trauma, unnecessary repeated hospitalisations, increases in economic cost and complications of stoma closure operation.9 For this reason, primary colon repair or resection with anastomosis after penetrating intestinal injuries have been suggested;10 however, colostomy should be preferred when there is gross peritoneal contamination, extensive colonic damage or shock at presentation.11 ,12 Although our primary treatment choices were primary repair procedures or anastomosis, we preferred resection and diversion stoma in some of the patients since our clinic is far enough from the battlefield and the transfer time was prolonged. The prolonged transfer time and insufficient resuscitation during this period caused gross peritoneal contamination and hypovolemic shock in some of the patients. In a patient admitted with haemorrhagic shock, we preferred a quick stoma instead of complete repair with or without anastomosis; widespread contamination was another reason for us to not perform primary repair. On the other hand, if the patients’ haemodynamics were stable with limited contamination, primary repair techniques were preferred.
Recently, it has been proposed that penetrating small intestinal injuries can be treated without any ileostomies or jejunostomies13 and most of the small intestine injuries in this study were treated with primary repair, but some with resection and anastomosis. Only a limited number of patients with distal ileum injuries required ileostomy or jejunostomy after resections.
The aim of the damage control principles is to take control of bleeding and contamination before the development of the lethal triad (metabolic acidosis, coagulopathy and hypothermia) in patients with severe abdominal trauma and rapid, often temporary, control of the major injuries is the first step of damage control surgery. After the patient becomes stable, an additional operation to repair the other injuries or definitively address the temporising measures should be performed.14 ,15 Of the 36 transferred patients, 12 were found to have retained abdominal packing suggestive of an earlier damage control approach; there were 18 cases with intestinal injuries and 8 of them had stomas. Two patients had undergone major vascular repair (for injuries of the inferior vena cava and left iliac artery, respectively). Among 80 primary surgery patients, we performed six liver packing procedures and three primary repairs for intestinal injuries to take control of peritoneal contamination. Also, 8 of 20 patients who were transferred to our clinic after receiving damage control procedures and 5 patients who had undergone damage control surgery in our clinic died.
During this period, six negative laparotomies were performed in patients who were transferred after previous surgery in Syria but arrived without any accompanying medical documentation. There were abdominal wall bullet or fragment entry wounds—all had abdominal pain, tenderness and rebound tenderness and three had haemorrhagic drain outputs. Although abdominal CT studies were normal, laparotomies were performed based on the clinical findings described. Negative laparotomy rates of up to 20% in war wounding are reported,16 but recently, advanced radiological techniques and even explorative laparoscopy can decrease numbers of false laparotomies.17–20 Laparoscopic evaluation after penetrating trauma is more cost effective than false laparotomy.21
The transfer time of injured cases is clearly related with complications and mortality. Rapid evacuation of the wounded cases reduces the mortality.6 ,22 The first 60 min after trauma is known as ‘golden hour’, and initial intervention in this period plays a critical role for the prognosis of the patient.23 However, the transfer time of the patients in our study was quite prolonged. The battlefields were often far from the border, and insufficient transport facilities from the battlefield to the Turkish border increased the transport time. Patients with hollow organ injuries were admitted with gross peritoneal contamination, and haemorrhagic shock was present in most of the patients with solid organ injuries. Septic complications were the leading cause of death. To reduce the transfer time in a war occurring in a nearby country, the civilian hospitals that are closer to the border should be transformed to military ones. During ‘August War 2008’ in Georgia, a civil hospital was transformed to a military one and provided surgery according to damage control principles.24 Transformation of a civilian hospital to a military one near to the border provides decreased transport time from battlefield to medical centre, correct triage for effective surgical treatment and initial intervention, if possible, in the golden hour.
Conclusion
War does not affect only the battlefield but also the neighbouring countries by virtue of military economic and medical support issues. Intestinal injuries are the most common intra-abdominal pathologies during wars. Delayed interventions are related with increased morbidity and mortality, and damage control surgery such as diversion stoma and abdominal packing becomes necessary. To avoid casualties, adequate medical organisation is required even if the war occurs in your neighbour. Some precautions should be taken such as transformation of nearby civilian hospitals to military ones to decrease transfer time and to perform adequate interventions. Trauma experienced surgeons should be staffed, complete drug and medical supplies provided and appropriate intensive care units should be designed in hospitals of border cities until the end of the war.
References
Footnotes
Contributors SA designed the study, searched the literature, collected, interpreted and analysed the data, wrote and revised the article. AA searched the literature interpreted and analysed the data, revised the article. IY collected and analysed the data, wrote and revised the article. MU searched the literature, interpreted and analysed the data, revised the article. CO, EK, AK and IP collected the data, searched the literature and revised the article. MT designed the study, collected and analysed the data and revised the article.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.