- Email: [email protected]
The spectrum and outcome of pregnant trauma patients in a metropolitan trauma service in South Africa
Accepted Manuscript Title: The spectrum and outcome of pregnant trauma patients in a metropolitan trauma service in South Africa Author: S. Wall F. Figueiredo G.L. Laing D.L. Clarke PII: DOI: Reference:
S0020-1383(14)00219-8 http://dx.doi.org/doi:10.1016/j.injury.2014.04.045 JINJ 5730
To appear in:
Injury, Int. J. Care Injured
Received date: Revised date: Accepted date:
30-12-2013 28-4-2014 30-4-2014
Please cite this article as: Wall S, Figueiredo F, Laing GL, Clarke DL, The spectrum and outcome of pregnant trauma patients in a metropolitan trauma service in South Africa, Injury (2014), http://dx.doi.org/10.1016/j.injury.2014.04.045 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The spectrum and outcome of pregnant trauma patients in a metropolitan trauma service in South Africa
Pietermaritzburg Metropolitan Trauma Service Pietermaritzburg Metropolitan Hospital Complex
cr
Department of General Surgery
ip t
Wall S, Figueiredo F, Laing GL, Clarke DL
Ac ce p
te
d
M
an
us
University of KwaZulu-Natal Nelson R Mandela School of Medicine
Page 1 of 13
An spectrum and outcome of trauma in pregnant patients in a metropolitan
Ac ce p
te
d
M
an
us
cr
ip t
trauma service in South Africa
Page 2 of 13
Introduction: Pregnant patients involved in trauma pose unique diagnostic and treatment challenges as the
ip t
physiological and anatomical changes associated with pregnancy, and the need to preserve fetal well-being, result in a number of nuances in the standard resuscitation algorithms. This clinical audit from a busy developing world trauma service describes the spectrum and
cr
outcome of pregnant trauma patients. Methods:
us
All pregnant patients presenting to the Pietermaritzburg Hospital Complex following trauma were included in the study. Data was retrieved both from trauma registry in Pietermaritzburg. The data was analyzed using descriptive statistics in a spreadsheet. The study ran from July
an
2011 to December 2013. Results:
M
During the thirty-month period under review, 1075 female trauma patients were admitted, 4% of these patients were pregnant (42/1075). The mean age of the patients in the study was 24.9 with an average gestation of 21.4 weeks. Blunt trauma accounted for the majority of the
d
trauma (57%). Trauma was intentional assault in 52% of the cases. Of the cases of assault,
te
81% of the time, the assailant was known to the victim and in the majority of cases (55%) the assailant was the patient’s intimate partner. Polytrauma predominated as the most common pattern of injury. Fetal death occurred in more than a third of cases (15/42). In 90% of the
Ac ce p
patients with an Injury Severity Score greater than fifteen, there was fetal death. Eighty-six percent (6/7) of the patients who required surgery had an unfavorable fetal outcome. In seventy-three percent of the cases of fetal death, the pregnancies were less the 28 weeks gestation.
Conclusion:
In an environment with high rates of interpersonal violence, the rate of trauma in pregnancy is not an uncommon occurrence. It is most commonly due to assault and the assailant is known to the victim in the majority of cases. Blunt trauma still predominate in this setting but there is a high incidence of penetrating trauma. Fetal mortality in this group is high and reflects the severity of the trauma experienced.
Page 3 of 13
ip t
Introduction The management of the pregnant trauma patient poses unique diagnostic and treatment
challenges as the physiological and anatomical changes associated with pregnancy and the
cr
need to preserve fetal well-being create a number of nuances in the standard resuscitation algorithms. ¹ Almost all textbooks and trauma courses devote a chapter to the issue of
managing the pregnant trauma patient; despite this there is a paucity of clinical reports on
us
trauma in pregnancy. ² This objective of this clinical audit from a busy developing world trauma service is to describe the spectrum and outcome of trauma in pregnancy in our environment and to attempt to develop guidelines and algorithms based on the data
an
presented.
M
Setting
The Pietermaritzburg Metropolitan Trauma Service (PMTS) was established in 2006 with the intention of providing comprehensive trauma care to the city of Pietermaritzburg and the
d
whole of western Kwa-Zulu Natal Province. For a number of pragmatic reasons, the PMTS could not adopt the concept of a trauma centre located in a single hospital and opted to
te
attempt to provide care across the metropolitan complex. Hence the PMTS is a service and not a center and this is reflected in it’s vision statement, “Taking Care of Pietermaritzburg”
Ac ce p
Trauma patients in Pietermaritzburg are directed to one of the three hospitals within the city depending on their geographical location. This model of delivering trauma care is somewhat unique and whilst philosophically attractive does present challenges in terms of the implementation of safe and pragmatic management algorithms. The PMTS aims to provide resources and expertise, as well as strategic and political leadership in trauma care to the city of Pietermaritzburg and the western rural health-districts of KZN. Pietermaritzburg is the capital of KZN and is the largest city in the western part of the province. It has a population of 1 001 000 people and is served by a tertiary hospital (Greys), a regional hospital (Edendale) and a district hospital (Northdale). There are three private hospitals in the city, which function entirely independently. Western KZN is a predominantly rural province with a population of two million people, and consists of four health districts. There are two other regional hospitals in Western KZN and 19 district hospitals. Resources The PMTS has six specialist surgeons. This comprises two registered sub-specialist trauma surgeons, two surgeons enrolled in sub-specialist trauma fellowship training and another two
Page 4 of 13
general surgeons with interests in trauma surgery. All trainees and junior staff work under the supervision of the specialist staff. At an operational level, the staff are sub-divided into three teams, one based at Greys hospital and two at Edendale hospital. Both hospitals have twenty-four hour availability of a computed tomography (CT) scanner, and have support from a radiology service. Each hospital provides a single emergency operative theatre, which
ip t
caters for the competitive needs of all surgical specialties. The complex has 10 high dependency unit (HDU) beds (3 adult and 7 pediatric) and 19 intensive care unit (ICU) beds (12 adult and 7 pediatric), which are shared by all the surgical disciplines. ICU beds are
cr
available at both Greys and Edendale hospitals. ICU patients are managed with the
Department of Critical Care in a “closed unit” fashion. There is a specialist lead obstetrical service in Pietermaritzburg, which covers all three hospitals in the complex. There is no onsite
us
neurosurgical service and all neurosurgery is performed in Durban 90 kilometers away. We attempt to follow ATLS guidelines for the management of all trauma patients. All female trauma victims of reproductive age undergo a routine pregnancy test and all pregnant trauma
an
patients are managed in conjunction with the staff of the obstetrical service.
M
The Pietermaritzburg Trauma Registry
The PMTS has mantained a paper based trauma registry since 2006 and implemented a comprehensive electronic surgical registry (ESR) in January 2012. Ethical approval to
d
construct and implement the ESR was obtained prior to its implementation (ethics number
te
BCA221/13 BREC UKZN).³ A standardized paper-based tick box style admission document is used for the admission of all trauma patients. The data is then extracted from this document and entered directly into the trauma registry at the time of discharge. Data is analyzed and
Ac ce p
validated by the system administrator on a weekly basis. Omitted data sets are identified and where possible completed from the discharge summary or patient records. Registry data is manually checked on a weekly basis against morning hand-over reports, and furthermore on a three monthly basis against hospital and city mortuary records. Methods
In addition to the ethics class approval for the ESR, (BCA221/13 BREC UKZN), this project was also covered by a separate ethics application BE 191/13. The study was conducted in the hospitals in the Pietermaritzburg Hospital Complex, which comprises Greys (tertiary), Edendale (regional) and Northdale (district) Hospitals. All pregnant patients admitted to one of these hospitals following trauma were included in this study. In addition to the data retrieved from the ESR each pregnant trauma patient was reviewed by the primary author and a data survey sheet completed. This sheet collected all the variables relevant to the study, namely mechanism of injury, pattern of injury, fetal outcome and in the case of assault, whether the assailant was the patient’s intimate partner or not. The data was analyzed using descriptive statistics in a spreadsheet. The study ran from July 2011 till December 2013.
Page 5 of 13
Results During the thirty-month period under review a total of forty-two pregnant trauma patients were seen in the three hospitals that comprise the Pietermaritzburg Metropolitan Complex. During this time we admitted 1075 female trauma patients and saw 3000 female trauma patients in the emergency rooms of the three hospitals of the PMB Complex. The incidence of trauma in
ip t
pregnancy in this study was four percent (42/1075). The mean age of the group was 24.9
years (Range 16 to 44 years). The average gestational age was 21.4 weeks. The gestational age was not recorded in 3 of the cases. Six patients were not aware that they were pregnant
cr
at the time of the trauma. Four of these were early in their first trimester and pregnancy was
diagnosed at the time of presentation for the trauma. The other two were each twenty weeks
us
pregnant and both denied being pregnant. There was an even distribution of gestational ages amongst the thirty-nine trauma victims with recorded gestation, the first, second and third trimeters comprising 12, 14 and 13 cases respectively. There was a single maternal death in flame burns and her death was expected.
M
Mechanism
an
the series. This patient sustained in excess of 70% mixed full thickness and partial thickness
Blunt trauma accounted for fifty-seven per cent of cases (24/42) and a combination of blunt and penetrating trauma accounted for another two cases. There were nine (21%) cases of
d
penetrating trauma. Burns accounted for a further five cases, there was a single case of corrosive ingestion and one case of attempted hanging. The trauma was intentional in 52% of
te
cases. The intentional cases included assault in the form of gunshot wounds (4), stab wounds (3) , blunt assault (9), a combination of blunt assault and stab wounds,(2) and assault with
Ac ce p
hot water (1) . There were two cases of self-inflicted trauma, an intentional corrosive ingestion and an attempted hanging. Of the assault cases the assailant was known to the victim in over eighty percent of cases and in fifty-five percent of cases, the assailant was the patient’s intimate partner. Non-intentional injury accounted for 48% of the trauma cases in this audit and road traffic crashes (RTC) accounted for 26 % (11) of the injuries in this group. There were three falls (3), one structural collapse (1) and four burns (4). There were two animal related injuries a snake-bite (1) and a dog bite (1). The accidental burns included a chemical burn (1) a burn from hot porridge (1) and flame burns (2). One of the flame burn victims was an eclamptic who suffered a seizure and fell into an open fire. Pattern of Trauma Polytrauma was the predominant pattern of injury in this study, accounting for forty percent (17/42). (Figure 1) Abdominal injuries were the next major injury, accounting for seventeen per cent (7/42). Head injuries accounted for just under 10% (4/42) of the cases. These were all mild head injuries with a GCS between 13-15. Eight of the patients had no significant surgical injuries, however three of these patients were admitted by the obstetrical service for
Page 6 of 13
vaginal bleeding. (Figure 1) The Injury Severity Score (ISS) ranged from 1 to 36. The average ISS was 8, with 15 of the patients receiving an ISS of 1. Ten of the patients had an ISS of greater than 15 indicating major trauma. The patient with the ISS of 36 was the patient who sustained in excess of 70% burns and subsequently demised. Laparotomy was required in five patients. In three cases for blunt trauma and in one case for a GSW and in the remaining
ip t
case for a stab wound of the abdomen. Fetal Outcome
cr
More than one third of the pregnancies (36%) were lost. Assault in the form of blunt assault,
gunshot wounds and assault with hot water, accounted for almost half of the fetal deaths (7 of the 15 fetal deaths). In the cases of fetal death, sixty percent (9/15) patients had an ISS of
us
greater than 15. In eight of the cases in which there was fetal demise, the mother sustained polytrauma. In four of the five cases where laparotomy was performed the fetus died. In two cases the mothers sustained no significant surgical trauma but progressed to spontaneous
an
miscarriage whilst admitted for observation by the obstetricians. Fetal demise also occurred in two patients with minor head injuries and in the one case of attempted hanging. All three of the fetuses were less than twenty-eight weeks of gestational age. Fetal demise occurred
M
more commonly in the first two trimesters (73%) than in the third trimester (27%). Discussion
d
Although trauma is listed as the leading cause of non–obstetric related maternal death, there
te
have not been many audits describing the spectrum and outcome of the entity in either the developed or developing world. This study shows that assault is the most common mechanism of injury in pregnant trauma patients in our environment and that trauma in
Ac ce p
pregnancy results in fetal loss in more than a third of cases. In a statewide review of trauma in pregnancy from California over a decade falls (N= 439) accounted for the biggest proportion of trauma in pregnancy followed by RTC’s (N=405) and then assaults (112) comprised just over eleven percent of all cases of trauma in pregnancy. Over a four-year
period the Virginia Commonwealth University Medical Center in, Richmond, Virginia admitted 142 pregnant trauma patients and discharged a further 152 from the emergency department. In that series the ratio of intentional to non-intentional injury was also lower than in our
series. Other American authors have however reported a much higher incidence of assault amongst pregnant trauma patients. Poole et al (1996) reported on a large series of pregnant trauma patients and the incidence of assault in that cohort was 31%. The high incidence of
assault in pregnancy that we observed in our study may be as a result of the low socioeconomic status of the patients in our setting with a resultant higher rate of alcohol abuse and violence. The high incidence of intimate partner related violence is a striking feature of our data as evidenced by the fact that the assailant was known to the victim in 81% of cases and was the intimate partner of the victim in 55% of cases. Blunt trauma predominates as the most common cause of injury in pregnant trauma patients in our study and this is in keeping
Page 7 of 13
with international trends, which attributes 90% of trauma in pregnancy to blunt mechanisms. The incidence of penetrating trauma in our cohort (21%) is higher than that reported in the literature where the rate ranges from three percent to ten percent.
A large population based study from California demonstrated that fetal
ip t
outcome is highly dependent on gestational age rather than and on the injury type, mechanism, or severity of the injury with the cohort less than twentyeight weeks gestation being most at risk. The authors postulated that this
cr
reflected poor fetal tolerance to maternal physiologic stress of trauma.
Although our cohort had a high rate of fetal loss, we do not have a large
us
enough sample to draw conclusions about the risk factors for fetal loss and further work is needed to accurately define these factors. The available
an
literature suggests that adverse fetal outcomes are related to the gestational age of the fetus as well as the severity of the injury and and the need for a surgical procedure during hospitalization. ¹ Our experience confirms this. In
M
ninety percent (10/11) of the cases of fetal death the mother had an ISS greater than 15 and seventy percent of fetal deaths occurred in the first two
d
trimesters. Furthermore direct major trauma to the abdomen and the need for surgery is associated with fetal loss in our setting. This is evidenced by the
te
fact that out of the seven pregnant trauma patients who required surgery, in the form of laparotomy, debridement of burn wounds or wash-out and fixation
Ac ce p
of a compound fracture and six (86%) had an unfavorable fetal outcome. (Table 1)
Conclusion
In an environment with high rates of interpersonal violence, trauma in pregnancy is not an uncommon occurrence. When it does occur in our setting it is mostly due to assault and the victim knows the assailant in the majority of cases. Although blunt trauma is the most common mechanism there is a high incidence of penetrating trauma and the rate of fetal loss is high. Further work is needed to accurately delineate the risk factors for fetal loss.
Page 8 of 13
References
1. Weintraub AY, Leron E, Mazor M. The pathophysiology of trauma in pregnancy: A review. J Mat-Fet and Neo Med. 2006 October; 19(10):601-605
ip t
2. American College of Surgeons Committee on Trauma (2012) Advanced trauma life support program for doctors, 9th edn. American College of Surgeons, Chicago
3. Laing GL, Bruce JL, Aldous C, Clarke DL. The design, construction and implementation of
cr
a computerised trauma registry in a developing South African metropolitan trauma service. Injury. 2014; 45:3-8
us
4. Corsi PR, Rasslan S, de Oliveira LB, Kronfly FS, Marinho VP. Trauma in pregnant women: analysis of maternal and fetal mortality. Injury. 1999; 30:239-243
5. El Kady D, Gilbert WM, Anderson J, Danielsen B, Towner D, Smith LH. Am J Obstet
an
Gynecol. 2004; 190:1661-1668
6. Aboutanos SZ, Aboutanos MB, Dompkowski D, Duane TM, Malhotra AK, Ivatury RR. Predictors of fetal outcome in pregnant trauma patients: a five year institutional review
M
7. Poole GV, Martin JN, Jr, Perry KG, Jr, Griswold JA, Lambert CJ, Rhoades RS. Trauma in pregnancy. Am J Obstet Gynecol. 1996; 174:1873-1877
8. Pearlman MD. Motor vehicle crashes, pregnancy loss and preterm labour. J Obstet. 1997;
d
57:127-132
9. Weiss HB: Pregnancy associated injury hospitalizations in Pennsylvania, 1995. Ann. 10.
te
Emerg. Med. 34(5), 626.636 (1999)
Theodorou DA, Velahos GC, Souter I, Chan LS, Vassiliu P, Tatevossian R et al.
Ac ce p
Fetal death after trauma in pregnancy. American Surgeon. 2000. 66:809-812
Page 9 of 13
Injury Severity Score
Surgery
cr
Injury sustained
ICU or High Care (HC) Required?
Ventilation Required?
ICU
Yes
< 12
PVA
Polytrauma: facial fractures, spinal fractures, rib fractures, pelvic fracture, compound tib-fib fracture
22
20
8
PVA
Grade 3 splenic injury, bilateral renal lacerations, pelvic fracture
18
No
Regional
HC
No
25
30
GSW Abdomen
Grade 4 liver laceration, renal injury, multiple placental infarcts
16
No
Tertiary
HC
No
29
8
Blunt Assualt
Mild head injury
1
No
Tertiary
No
No
29
< 12
GSW Face Pharyngeal injury, hypopharyngeal injury, left external carotid transection
24
Yes - Neck exploration
Regional
ICU
Yes
16
20
Hot Water Burns
36% burns to face, neck, chest, left arm and abdomen
9
Yes - Wound Regional debridement
Burns HC
No
28
32
GSW Back
Left internal iliac artery injury, left external iliac vein injury - resulted in left above knee amputation
16
ICU
Yes
34
20
Flame & Hot Water Burns
70% burns to face, chest, abdomen , arms and legs
36
Yes Regional Laparotomy + Caesarean Section No Regional
Burns HC
No - not a candidate for ventilation
23
24
MVA
Blunt chest & abdominal trauma: ruptured uterus
20
Yes Laparotomy
Tertiary
ICU
Yes
31
22
MVA
Only complaint: ↓ fetal movement
1
No
District
No
No
d
M
an
39
Ac c
Yes Fixation of fractures
Level Hospital where treated Tertiary
ep te
us
Mater Gestation Mechanism nal in weeks Age
ip t
Table
Page 10 of 13
ip t
Injury Severity Score
Surgery
ICU or High Care (HC) Required?
Ventilation Required?
No
Level Hospital where treated Tertiary
ICU
Yes
1
No
Regional
No
No
cr
Injury sustained
us
Mater Gestation Mechanism nal in weeks Age 30
Blunt Assualt
Multiple spinal injuries, base of skull fracture, intracranial haemorrhage, abruptio placenta
18
26
18
Blunt Assualt
No surgical injuries, spontaneous miscarriage 2 days post-assault
17
22
Eclamptic & Fall from stretcher
Mild head injury, gross facial swelling
4
No
Regional
Obstetric HC
No
33
38
MVA
Crush syndrome with acute kidney injury, grade 3b abruptio placenta
12
Regional
Obstetric HC
No
20
12
Hanging & Overdose
Hypoxic brain injury, initial GCS 9/15 improve to 12/15
Yes Caesarean Section No
Regional
ICU
Yes
ep te
d
M
an
26
25
Ac c
Table 1: Profile of Patients with adverse fetal outcomes.
Page 11 of 13
ip t
Figure
Ac ce pt e
d
M
an
us
cr
Figure 1
Page 12 of 13
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1 - Pattern of Injury.doc
Page 13 of 13
S0020-1383(14)00219-8 http://dx.doi.org/doi:10.1016/j.injury.2014.04.045 JINJ 5730
To appear in:
Injury, Int. J. Care Injured
Received date: Revised date: Accepted date:
30-12-2013 28-4-2014 30-4-2014
Please cite this article as: Wall S, Figueiredo F, Laing GL, Clarke DL, The spectrum and outcome of pregnant trauma patients in a metropolitan trauma service in South Africa, Injury (2014), http://dx.doi.org/10.1016/j.injury.2014.04.045 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The spectrum and outcome of pregnant trauma patients in a metropolitan trauma service in South Africa
Pietermaritzburg Metropolitan Trauma Service Pietermaritzburg Metropolitan Hospital Complex
cr
Department of General Surgery
ip t
Wall S, Figueiredo F, Laing GL, Clarke DL
Ac ce p
te
d
M
an
us
University of KwaZulu-Natal Nelson R Mandela School of Medicine
Page 1 of 13
An spectrum and outcome of trauma in pregnant patients in a metropolitan
Ac ce p
te
d
M
an
us
cr
ip t
trauma service in South Africa
Page 2 of 13
Introduction: Pregnant patients involved in trauma pose unique diagnostic and treatment challenges as the
ip t
physiological and anatomical changes associated with pregnancy, and the need to preserve fetal well-being, result in a number of nuances in the standard resuscitation algorithms. This clinical audit from a busy developing world trauma service describes the spectrum and
cr
outcome of pregnant trauma patients. Methods:
us
All pregnant patients presenting to the Pietermaritzburg Hospital Complex following trauma were included in the study. Data was retrieved both from trauma registry in Pietermaritzburg. The data was analyzed using descriptive statistics in a spreadsheet. The study ran from July
an
2011 to December 2013. Results:
M
During the thirty-month period under review, 1075 female trauma patients were admitted, 4% of these patients were pregnant (42/1075). The mean age of the patients in the study was 24.9 with an average gestation of 21.4 weeks. Blunt trauma accounted for the majority of the
d
trauma (57%). Trauma was intentional assault in 52% of the cases. Of the cases of assault,
te
81% of the time, the assailant was known to the victim and in the majority of cases (55%) the assailant was the patient’s intimate partner. Polytrauma predominated as the most common pattern of injury. Fetal death occurred in more than a third of cases (15/42). In 90% of the
Ac ce p
patients with an Injury Severity Score greater than fifteen, there was fetal death. Eighty-six percent (6/7) of the patients who required surgery had an unfavorable fetal outcome. In seventy-three percent of the cases of fetal death, the pregnancies were less the 28 weeks gestation.
Conclusion:
In an environment with high rates of interpersonal violence, the rate of trauma in pregnancy is not an uncommon occurrence. It is most commonly due to assault and the assailant is known to the victim in the majority of cases. Blunt trauma still predominate in this setting but there is a high incidence of penetrating trauma. Fetal mortality in this group is high and reflects the severity of the trauma experienced.
Page 3 of 13
ip t
Introduction The management of the pregnant trauma patient poses unique diagnostic and treatment
challenges as the physiological and anatomical changes associated with pregnancy and the
cr
need to preserve fetal well-being create a number of nuances in the standard resuscitation algorithms. ¹ Almost all textbooks and trauma courses devote a chapter to the issue of
managing the pregnant trauma patient; despite this there is a paucity of clinical reports on
us
trauma in pregnancy. ² This objective of this clinical audit from a busy developing world trauma service is to describe the spectrum and outcome of trauma in pregnancy in our environment and to attempt to develop guidelines and algorithms based on the data
an
presented.
M
Setting
The Pietermaritzburg Metropolitan Trauma Service (PMTS) was established in 2006 with the intention of providing comprehensive trauma care to the city of Pietermaritzburg and the
d
whole of western Kwa-Zulu Natal Province. For a number of pragmatic reasons, the PMTS could not adopt the concept of a trauma centre located in a single hospital and opted to
te
attempt to provide care across the metropolitan complex. Hence the PMTS is a service and not a center and this is reflected in it’s vision statement, “Taking Care of Pietermaritzburg”
Ac ce p
Trauma patients in Pietermaritzburg are directed to one of the three hospitals within the city depending on their geographical location. This model of delivering trauma care is somewhat unique and whilst philosophically attractive does present challenges in terms of the implementation of safe and pragmatic management algorithms. The PMTS aims to provide resources and expertise, as well as strategic and political leadership in trauma care to the city of Pietermaritzburg and the western rural health-districts of KZN. Pietermaritzburg is the capital of KZN and is the largest city in the western part of the province. It has a population of 1 001 000 people and is served by a tertiary hospital (Greys), a regional hospital (Edendale) and a district hospital (Northdale). There are three private hospitals in the city, which function entirely independently. Western KZN is a predominantly rural province with a population of two million people, and consists of four health districts. There are two other regional hospitals in Western KZN and 19 district hospitals. Resources The PMTS has six specialist surgeons. This comprises two registered sub-specialist trauma surgeons, two surgeons enrolled in sub-specialist trauma fellowship training and another two
Page 4 of 13
general surgeons with interests in trauma surgery. All trainees and junior staff work under the supervision of the specialist staff. At an operational level, the staff are sub-divided into three teams, one based at Greys hospital and two at Edendale hospital. Both hospitals have twenty-four hour availability of a computed tomography (CT) scanner, and have support from a radiology service. Each hospital provides a single emergency operative theatre, which
ip t
caters for the competitive needs of all surgical specialties. The complex has 10 high dependency unit (HDU) beds (3 adult and 7 pediatric) and 19 intensive care unit (ICU) beds (12 adult and 7 pediatric), which are shared by all the surgical disciplines. ICU beds are
cr
available at both Greys and Edendale hospitals. ICU patients are managed with the
Department of Critical Care in a “closed unit” fashion. There is a specialist lead obstetrical service in Pietermaritzburg, which covers all three hospitals in the complex. There is no onsite
us
neurosurgical service and all neurosurgery is performed in Durban 90 kilometers away. We attempt to follow ATLS guidelines for the management of all trauma patients. All female trauma victims of reproductive age undergo a routine pregnancy test and all pregnant trauma
an
patients are managed in conjunction with the staff of the obstetrical service.
M
The Pietermaritzburg Trauma Registry
The PMTS has mantained a paper based trauma registry since 2006 and implemented a comprehensive electronic surgical registry (ESR) in January 2012. Ethical approval to
d
construct and implement the ESR was obtained prior to its implementation (ethics number
te
BCA221/13 BREC UKZN).³ A standardized paper-based tick box style admission document is used for the admission of all trauma patients. The data is then extracted from this document and entered directly into the trauma registry at the time of discharge. Data is analyzed and
Ac ce p
validated by the system administrator on a weekly basis. Omitted data sets are identified and where possible completed from the discharge summary or patient records. Registry data is manually checked on a weekly basis against morning hand-over reports, and furthermore on a three monthly basis against hospital and city mortuary records. Methods
In addition to the ethics class approval for the ESR, (BCA221/13 BREC UKZN), this project was also covered by a separate ethics application BE 191/13. The study was conducted in the hospitals in the Pietermaritzburg Hospital Complex, which comprises Greys (tertiary), Edendale (regional) and Northdale (district) Hospitals. All pregnant patients admitted to one of these hospitals following trauma were included in this study. In addition to the data retrieved from the ESR each pregnant trauma patient was reviewed by the primary author and a data survey sheet completed. This sheet collected all the variables relevant to the study, namely mechanism of injury, pattern of injury, fetal outcome and in the case of assault, whether the assailant was the patient’s intimate partner or not. The data was analyzed using descriptive statistics in a spreadsheet. The study ran from July 2011 till December 2013.
Page 5 of 13
Results During the thirty-month period under review a total of forty-two pregnant trauma patients were seen in the three hospitals that comprise the Pietermaritzburg Metropolitan Complex. During this time we admitted 1075 female trauma patients and saw 3000 female trauma patients in the emergency rooms of the three hospitals of the PMB Complex. The incidence of trauma in
ip t
pregnancy in this study was four percent (42/1075). The mean age of the group was 24.9
years (Range 16 to 44 years). The average gestational age was 21.4 weeks. The gestational age was not recorded in 3 of the cases. Six patients were not aware that they were pregnant
cr
at the time of the trauma. Four of these were early in their first trimester and pregnancy was
diagnosed at the time of presentation for the trauma. The other two were each twenty weeks
us
pregnant and both denied being pregnant. There was an even distribution of gestational ages amongst the thirty-nine trauma victims with recorded gestation, the first, second and third trimeters comprising 12, 14 and 13 cases respectively. There was a single maternal death in flame burns and her death was expected.
M
Mechanism
an
the series. This patient sustained in excess of 70% mixed full thickness and partial thickness
Blunt trauma accounted for fifty-seven per cent of cases (24/42) and a combination of blunt and penetrating trauma accounted for another two cases. There were nine (21%) cases of
d
penetrating trauma. Burns accounted for a further five cases, there was a single case of corrosive ingestion and one case of attempted hanging. The trauma was intentional in 52% of
te
cases. The intentional cases included assault in the form of gunshot wounds (4), stab wounds (3) , blunt assault (9), a combination of blunt assault and stab wounds,(2) and assault with
Ac ce p
hot water (1) . There were two cases of self-inflicted trauma, an intentional corrosive ingestion and an attempted hanging. Of the assault cases the assailant was known to the victim in over eighty percent of cases and in fifty-five percent of cases, the assailant was the patient’s intimate partner. Non-intentional injury accounted for 48% of the trauma cases in this audit and road traffic crashes (RTC) accounted for 26 % (11) of the injuries in this group. There were three falls (3), one structural collapse (1) and four burns (4). There were two animal related injuries a snake-bite (1) and a dog bite (1). The accidental burns included a chemical burn (1) a burn from hot porridge (1) and flame burns (2). One of the flame burn victims was an eclamptic who suffered a seizure and fell into an open fire. Pattern of Trauma Polytrauma was the predominant pattern of injury in this study, accounting for forty percent (17/42). (Figure 1) Abdominal injuries were the next major injury, accounting for seventeen per cent (7/42). Head injuries accounted for just under 10% (4/42) of the cases. These were all mild head injuries with a GCS between 13-15. Eight of the patients had no significant surgical injuries, however three of these patients were admitted by the obstetrical service for
Page 6 of 13
vaginal bleeding. (Figure 1) The Injury Severity Score (ISS) ranged from 1 to 36. The average ISS was 8, with 15 of the patients receiving an ISS of 1. Ten of the patients had an ISS of greater than 15 indicating major trauma. The patient with the ISS of 36 was the patient who sustained in excess of 70% burns and subsequently demised. Laparotomy was required in five patients. In three cases for blunt trauma and in one case for a GSW and in the remaining
ip t
case for a stab wound of the abdomen. Fetal Outcome
cr
More than one third of the pregnancies (36%) were lost. Assault in the form of blunt assault,
gunshot wounds and assault with hot water, accounted for almost half of the fetal deaths (7 of the 15 fetal deaths). In the cases of fetal death, sixty percent (9/15) patients had an ISS of
us
greater than 15. In eight of the cases in which there was fetal demise, the mother sustained polytrauma. In four of the five cases where laparotomy was performed the fetus died. In two cases the mothers sustained no significant surgical trauma but progressed to spontaneous
an
miscarriage whilst admitted for observation by the obstetricians. Fetal demise also occurred in two patients with minor head injuries and in the one case of attempted hanging. All three of the fetuses were less than twenty-eight weeks of gestational age. Fetal demise occurred
M
more commonly in the first two trimesters (73%) than in the third trimester (27%). Discussion
d
Although trauma is listed as the leading cause of non–obstetric related maternal death, there
te
have not been many audits describing the spectrum and outcome of the entity in either the developed or developing world. This study shows that assault is the most common mechanism of injury in pregnant trauma patients in our environment and that trauma in
Ac ce p
pregnancy results in fetal loss in more than a third of cases. In a statewide review of trauma in pregnancy from California over a decade falls (N= 439) accounted for the biggest proportion of trauma in pregnancy followed by RTC’s (N=405) and then assaults (112) comprised just over eleven percent of all cases of trauma in pregnancy. Over a four-year
period the Virginia Commonwealth University Medical Center in, Richmond, Virginia admitted 142 pregnant trauma patients and discharged a further 152 from the emergency department. In that series the ratio of intentional to non-intentional injury was also lower than in our
series. Other American authors have however reported a much higher incidence of assault amongst pregnant trauma patients. Poole et al (1996) reported on a large series of pregnant trauma patients and the incidence of assault in that cohort was 31%. The high incidence of
assault in pregnancy that we observed in our study may be as a result of the low socioeconomic status of the patients in our setting with a resultant higher rate of alcohol abuse and violence. The high incidence of intimate partner related violence is a striking feature of our data as evidenced by the fact that the assailant was known to the victim in 81% of cases and was the intimate partner of the victim in 55% of cases. Blunt trauma predominates as the most common cause of injury in pregnant trauma patients in our study and this is in keeping
Page 7 of 13
with international trends, which attributes 90% of trauma in pregnancy to blunt mechanisms. The incidence of penetrating trauma in our cohort (21%) is higher than that reported in the literature where the rate ranges from three percent to ten percent.
A large population based study from California demonstrated that fetal
ip t
outcome is highly dependent on gestational age rather than and on the injury type, mechanism, or severity of the injury with the cohort less than twentyeight weeks gestation being most at risk. The authors postulated that this
cr
reflected poor fetal tolerance to maternal physiologic stress of trauma.
Although our cohort had a high rate of fetal loss, we do not have a large
us
enough sample to draw conclusions about the risk factors for fetal loss and further work is needed to accurately define these factors. The available
an
literature suggests that adverse fetal outcomes are related to the gestational age of the fetus as well as the severity of the injury and and the need for a surgical procedure during hospitalization. ¹ Our experience confirms this. In
M
ninety percent (10/11) of the cases of fetal death the mother had an ISS greater than 15 and seventy percent of fetal deaths occurred in the first two
d
trimesters. Furthermore direct major trauma to the abdomen and the need for surgery is associated with fetal loss in our setting. This is evidenced by the
te
fact that out of the seven pregnant trauma patients who required surgery, in the form of laparotomy, debridement of burn wounds or wash-out and fixation
Ac ce p
of a compound fracture and six (86%) had an unfavorable fetal outcome. (Table 1)
Conclusion
In an environment with high rates of interpersonal violence, trauma in pregnancy is not an uncommon occurrence. When it does occur in our setting it is mostly due to assault and the victim knows the assailant in the majority of cases. Although blunt trauma is the most common mechanism there is a high incidence of penetrating trauma and the rate of fetal loss is high. Further work is needed to accurately delineate the risk factors for fetal loss.
Page 8 of 13
References
1. Weintraub AY, Leron E, Mazor M. The pathophysiology of trauma in pregnancy: A review. J Mat-Fet and Neo Med. 2006 October; 19(10):601-605
ip t
2. American College of Surgeons Committee on Trauma (2012) Advanced trauma life support program for doctors, 9th edn. American College of Surgeons, Chicago
3. Laing GL, Bruce JL, Aldous C, Clarke DL. The design, construction and implementation of
cr
a computerised trauma registry in a developing South African metropolitan trauma service. Injury. 2014; 45:3-8
us
4. Corsi PR, Rasslan S, de Oliveira LB, Kronfly FS, Marinho VP. Trauma in pregnant women: analysis of maternal and fetal mortality. Injury. 1999; 30:239-243
5. El Kady D, Gilbert WM, Anderson J, Danielsen B, Towner D, Smith LH. Am J Obstet
an
Gynecol. 2004; 190:1661-1668
6. Aboutanos SZ, Aboutanos MB, Dompkowski D, Duane TM, Malhotra AK, Ivatury RR. Predictors of fetal outcome in pregnant trauma patients: a five year institutional review
M
7. Poole GV, Martin JN, Jr, Perry KG, Jr, Griswold JA, Lambert CJ, Rhoades RS. Trauma in pregnancy. Am J Obstet Gynecol. 1996; 174:1873-1877
8. Pearlman MD. Motor vehicle crashes, pregnancy loss and preterm labour. J Obstet. 1997;
d
57:127-132
9. Weiss HB: Pregnancy associated injury hospitalizations in Pennsylvania, 1995. Ann. 10.
te
Emerg. Med. 34(5), 626.636 (1999)
Theodorou DA, Velahos GC, Souter I, Chan LS, Vassiliu P, Tatevossian R et al.
Ac ce p
Fetal death after trauma in pregnancy. American Surgeon. 2000. 66:809-812
Page 9 of 13
Injury Severity Score
Surgery
cr
Injury sustained
ICU or High Care (HC) Required?
Ventilation Required?
ICU
Yes
< 12
PVA
Polytrauma: facial fractures, spinal fractures, rib fractures, pelvic fracture, compound tib-fib fracture
22
20
8
PVA
Grade 3 splenic injury, bilateral renal lacerations, pelvic fracture
18
No
Regional
HC
No
25
30
GSW Abdomen
Grade 4 liver laceration, renal injury, multiple placental infarcts
16
No
Tertiary
HC
No
29
8
Blunt Assualt
Mild head injury
1
No
Tertiary
No
No
29
< 12
GSW Face Pharyngeal injury, hypopharyngeal injury, left external carotid transection
24
Yes - Neck exploration
Regional
ICU
Yes
16
20
Hot Water Burns
36% burns to face, neck, chest, left arm and abdomen
9
Yes - Wound Regional debridement
Burns HC
No
28
32
GSW Back
Left internal iliac artery injury, left external iliac vein injury - resulted in left above knee amputation
16
ICU
Yes
34
20
Flame & Hot Water Burns
70% burns to face, chest, abdomen , arms and legs
36
Yes Regional Laparotomy + Caesarean Section No Regional
Burns HC
No - not a candidate for ventilation
23
24
MVA
Blunt chest & abdominal trauma: ruptured uterus
20
Yes Laparotomy
Tertiary
ICU
Yes
31
22
MVA
Only complaint: ↓ fetal movement
1
No
District
No
No
d
M
an
39
Ac c
Yes Fixation of fractures
Level Hospital where treated Tertiary
ep te
us
Mater Gestation Mechanism nal in weeks Age
ip t
Table
Page 10 of 13
ip t
Injury Severity Score
Surgery
ICU or High Care (HC) Required?
Ventilation Required?
No
Level Hospital where treated Tertiary
ICU
Yes
1
No
Regional
No
No
cr
Injury sustained
us
Mater Gestation Mechanism nal in weeks Age 30
Blunt Assualt
Multiple spinal injuries, base of skull fracture, intracranial haemorrhage, abruptio placenta
18
26
18
Blunt Assualt
No surgical injuries, spontaneous miscarriage 2 days post-assault
17
22
Eclamptic & Fall from stretcher
Mild head injury, gross facial swelling
4
No
Regional
Obstetric HC
No
33
38
MVA
Crush syndrome with acute kidney injury, grade 3b abruptio placenta
12
Regional
Obstetric HC
No
20
12
Hanging & Overdose
Hypoxic brain injury, initial GCS 9/15 improve to 12/15
Yes Caesarean Section No
Regional
ICU
Yes
ep te
d
M
an
26
25
Ac c
Table 1: Profile of Patients with adverse fetal outcomes.
Page 11 of 13
ip t
Figure
Ac ce pt e
d
M
an
us
cr
Figure 1
Page 12 of 13
Ac ce p
te
d
M
an
us
cr
ip t
Figure 1 - Pattern of Injury.doc
Page 13 of 13