Article Text
PDF
Abstract
Background We conducted a monthly epidemiological survey in Germany to detect the prevalence of volvulus without malrotation (VWM) in very low birthweight (VLBW) infants and to identify factors for a better distinction between this rare and life-threatening event and other acute abdominal diseases in preterm infants.
Methods Throughout 2014 and 2015, every paediatric department in Germany was asked to report cases of VWM in infants with birth weights <1500 g to the Surveillance Unit for Rare Paediatric Conditions in Germany. Hospitals reporting a case were asked to return an anonymised questionnaire and discharge letter.
Results Of 36 cases reported, detailed information was submitted on 29, with 26 meeting entry criteria. With 9896 and 10 140 VLBW infants born in Germany in 2014 and 2015, respectively, we estimated a prevalence of 1.52/1000 VLBW infants for 2014 and 1.08/1000 in 2015. No specific early symptom could be determined. 10% died, and 21% of infants developed short bowel syndrome. Calculated from all verified VWM arise death in 8% and a short bowel syndrome in 15%.
Conclusion VWM is a rare source of acute abdomen in VLBW infants. No specific signs and symptoms potentially facilitating an early recognition of VWM could be found from this survey. Because the rates of death and short bowel syndrome are high, VWM should always be considered early in an acute abdomen in a VLBW infant.
- acute abdomen in preterm VLBW infants
- malrotation without volvulus
- epidemiology
- mortality
- injury prevention
Statistics from Altmetric.com
- acute abdomen in preterm VLBW infants
- malrotation without volvulus
- epidemiology
- mortality
- injury prevention
What is already known on this topic?
Volvulus without malrotation (VWM) is a rare and life-threatening disease, needing immediate surgical intervention to prevent mortality and short bowel syndrome.
What this study adds?
This study shows that VWM occurs in 1–1.5/1000 infants <1500 g, more often than expected.
Differentiation to other abdominal diseases remains difficult and challenging.
Introduction
Volvulus without malrotation (VWM) represents a life-threatening event in premature infants. Case series describe its occurrence typically in very low birthweight (VLBW) infants and several weeks after birth.1 2 VWM generally presents as an acute abdomen with a rapid and dramatic deterioration of the infant’s general condition.3 Especially in the vulnerable population of very preterm infants, an early diagnosis, that is, before the occurrence of irreversible intestinal ischaemia, is difficult due to a rapid progression of ischaemic bowel damage and non-characteristic early clinical signs or imaging findings.4 Furthermore, VWM may be mistaken for the more common entity of necrotising enterocolitis (NEC) in VLBW infants.5 6 Conservative treatment for suspected NEC delays surgical intervention6 thereby aggravating morbidity and mortality.
A VWM requires immediate surgical treatment to restore intestinal blood supply and to avoid irreversible ischaemic bowel damage. Ischaemia of the twisted bowel is supposed to develop more rapidly in VWM because of the normal position and fixation of the uninvolved colon, whereas in cases of volvulus with malrotation and a mobile caecum, the colon may attenuate ischaemia and sequelae of the small bowel torsion.3 Both incidence and aetiology of VWM are unknown.3 In preterm infants, respiratory support with continuous positive airway pressure (CPAP) as well as abdominal massage and pelvic rotation have been suspected as risk factors.1 2
We performed a prospective epidemiological study in Germany (1) to determine the incidence of VWM in VLBW infants, (2) to detect potential ‘red flag’ signs and symptoms of this disease, (3) to gather information on risk factors and (4) to potentially develop preventive strategies.
Patients and methods
As part of the Surveillance Unit for Rare Paediatric Conditions in Germany (ESPED), all paediatric departments (459 in 2014 and 409 in 2015) received monthly reporting cards asking them about new admissions of preterm VLBW infants with VWM admitted between January 2014 and December 2015. Reports on the mailing card prompted immediate mailing of an anonymised three-page questionnaire. Electronic reminders were sent to all non-responders. In case of persistently missing completed questionnaires, the study centre also contacted the local person responsible for the ESPED collaboration to maximise response rates. Captured were all preterm infants with a birth weight of <1500 g who had suffered an intestinal VWM. VWM had been defined as any finding of volvulus on laparotomy, where the duodenal rotation/fixation was confirmed as normal (also known as ‘segmental volvulus’). In general, reporting completeness of the ESPED system exceeds 95%.
Data collection
The anonymised three-page questionnaire designed by our group asked for basic demographic data, information about pregnancy, perinatal medical history and the clinical course during the postnatal hospitalisation. The latter included details on clinical signs, diagnostic imaging and laboratory findings on the day of volvulus presentation and on parameters of gastrointestinal function and enteral nutrition, manipulations intended to promote bowel function, as well as respiratory support received prior to the event. Furthermore, details on intraoperative findings, including the extent of bowel resection (if any) and the postoperative outcome were obtained, as was an anonymised copy of the infant’s discharge letter. Based on these data, duplicate cases or those not meeting inclusion criteria were excluded.
Statistical analyses
As this is an explorative study without a primary hypothesis, no sample size calculation was done. Descriptive statistics were applied to characterise the study population. Data are presented as median (range).
Results
Between January 2014 and December 2015, a total of 36 cases of VWM in VLBW infants were reported to the study centre; detailed information was supplied in 29 cases, yielding a response rate of 81%. Unfortunately, several respondents did not provide answers to all questionnaire items. Response rates for single questionnaire items ranged from 40% to 100%
Of the 29 cases with detailed information, 26 could be verified (15 in 2014 and 11 in 2015) while three with diagnosis other than VWM were excluded.
Given a birth rate of 9896 preterm VLBW infants in 2014 and 10 140 in 2015 in Germany (according to the German Federal Statistical Office) and assuming complete reporting, an incidence of intestinal VWM of 1.29/1000 VLBW infants (95% CI 1.08 to 1.51) was estimated: 1.52/1000 in 2014 and 1.08/1000 in 2015. All diagnoses of VWM were made on surgery. Eighty-one per cent (21/26) of the affected infants were predominantly of extremely low gestational age (<28 weeks); they presented at a median postnatal age of 27 days. Median Apgar values at 1/5/10 min were 6/7/9; median umbilical artery pH was 7.34. Five infants were twins; 54% were girls. Detailed demographic data are shown in table 1.
Demographic data
Prenatal ultrasound and neonatal data
Prenatal ultrasound findings were reported as suspicious in only two cases who had dilated echogenic bowels and absent peristalsis. One of these infants also had a whirlpool sign and underwent laparotomy immediately after birth, on which 32 cm of necrotic bowel was removed. Although she subsequently developed short bowel syndrome (SBS), she could be discharged on full enteral feeds. The other infant had laparotomy 13 days postnatally, 2 days after first becoming symptomatic for VWM. No information on intraoperative findings or the postoperative course was supplied. No polyhydramnions or ascites was reported antenatally, but four had pathological Doppler signals.
Enteral intake and stooling before diagnosis of VWM
Information on enteral feeding before the event was available for 22 infants, half of which were solely fed with mother’s milk; in five infants, a fortifier had been added. Five infants were only formula fed, while six had received both breast milk and formula. The median time interval from birth until full enteral feeding was achieved was 20 days (range 6–60; 16 responses).
The time interval from birth until the last meconium-like stool was passed was 7 days (range 2–20). Tolerance to enteral feeding was unremarkable before the event, except for four patients who had bilious and two patients who had bloody gastric residues reported within the last 5 days before the event. Support given for stooling in the last 5 days before the event had been reported for 15 infants included enemas (n=14), abdominal massage (n=12) and/or a rectal tube (n=6). Only six patients were reported to defecate completely spontaneously. One patient was operated directly after birth without any support attempts. For four patients, information is lacking.
Respiratory support
At time of VWM diagnosis, most infants had received support via nasal CPAP. The highest positive end expiratory pressure (PEEP) used ranged from 4 to 8 cm H2O, with a median of 6 cm H2O (table 2).
Type and proportion of respiratory support before diagnosis of VWM
Clinical and radiological signs
Clinical signs at presentation were dominated by rather unspecific findings except that they pointed to a gastrointestinal problem (table 3).
Main symptoms on the day of VWM diagnosis
Radiological and ultrasound findings are described in table 4. The most common finding on ultrasound as well as on plain radiographs was dilated bowel loops followed by a suspicious configuration of bowel loops on plain radiographs. No specific imaging findings were recognised, and a whirlpool sign was hardly ever reported.
Abdominal imaging findings on the day of VWM diagnosis
Time to laparotomy, intraoperative and postoperative findings
In eight infants, laparotomy was performed on the day when symptoms of VWM were first noticed, in seven infants laparotomy was done on day 1, in four infants on day 2, in two infants on day 4 and in one infant each on days 5, 6 and 22, respectively, after onset of clinical signs. In two infants the timing of laparotomy in relation to onset of clinical signs was not reported.
Two infants had received a laparotomy prior to VWM, one for stomach perforation on postnatal day 8, and 90 days later the laparotomy was carried out because of a VWM.
The other underwent an explorative laparotomy because of an ileus symptomatology on the 55th day of life, and 94 days later needed a laparotomy because of VWM.
In 14 of 24 cases with information provided, volvulus had been suspected preoperatively; in the remaining infants, the diagnosis was only made intraoperatively.
Intraoperative findings besides volvulus were an ovarian cyst, a microcolon, a missing fixation with a mobile colon and Meckel’s diverticulum and a segmental intestinal stenosis of 2 mm in length (in one infant each).
Eighteen of 25 infants (72%) had a bowel resection ranging from 4 cm to resection of the entire small intestine (median: 23 cm); information on bowel resection was missing in one infant.
Eleven infants required a second laparotomy, two had two relaparotomies. Although a second laparotomy could be part of a surgical strategy to allow as much ischaemic bowel as possible to recover before definite decision for resection, this strategy was not applied in this cohort. Eighteen patients survived, while two died shortly after laparotomy (six infants with missing data). Four infants developed a SBS (SBS was defined as discharge with partial parental nutrition); one of these was on total parenteral nutrition at discharge, two were on full enteral feedings and in one infant information on this item is missing.
Discussion
We wanted to determine the prevalence of VWM in Germany and identify specific signs that would help to distinguish VWM from more common conditions like NEC or spontaneous intestinal perforation in preterm infants. We were notified of 26 confirmed cases over a 2-year study period, yielding a prevalence of about 1.3/1000 VLBW infants in Germany. To our best knowledge, this is the first report on population-based prevalence data of VWM in VLBW infants. For comparison, a British neonatal surgical service reported only one case of VWM in a series of 57 emergency laparotomies in extremely low gestational age neonates, indicating that other causes of acute abdomen dominate by far.7
In contrast to previous case series, VWM occurred equally often in girls and boys, and comparatively late after birth: median age here was 27 days, whereas in the other report, the median age was about 20 days.1–6 8–10 Yet, patients in this survey were born at a lower gestational age (25.9 weeks). The median gestational age in previously reported case series was about 28 weeks.1–6 8–10 Because this survey focused on VLBW infants, our study group was more premature. Prenatal manifestation of volvulus, as here reported for two infants, has been reported before.9 The diagnosis of VWM was made during laparotomy 27 (0–149) days after birth. In this survey, there was no apparent relationship between postnatal age at surgery and severity of outcome. Infants who died had onset of clinical signs on postnatal day 82 (64–99), and all had laparotomy on the day of initial presentation.
Infants with delayed surgery (first surgical intervention more than 1 day after onset of symptoms) had favourable outcomes (all survived without (temporary) SBS). However, this should not be interpreted as an indication to postpone surgical intervention; to the contrary, infants with delayed surgical interventions probably had less severe twisting of the intestine and hence preserved blood flow.
Clinical presentation was that of an acute abdomen without any VWM-specific signs. Similarly, recent case reports did not identify specific clinical sign that could help identify VWM. In our study, the main clinical symptoms were acute abdominal distension (69%), acute deterioration of the infants’ general condition (65%) and suddenly increased bilious gastric residuals (38%).
Respiratory support via CPAP has been discussed as a risk factor for VWM.9 Fifty-six per cent of the infants in this cohort had respiratory support with CPAP the day before VWM manifestation, but at PEEP levels well within the expected range, given that most infants were born <32 weeks and only about 3–4 weeks old. Due to the lack of a matched control group in this study, it remains uncertain whether respiratory support (CPAP in 56% of patients) or any other respiratory support are indeed risk factors or just normal for this patient population. Our own series of VWM previously published along with control infants matched for gestational age at birth, sex, birth year and postnatal age at birth, however, indicated that more intense positive pressure support might contribute to VWM.10
As also reported by others,3 4 neither radiological imaging nor ultrasound examination was successful in detecting volvulus-specific features. Intramural gas was reported in three cases. A whirlpool sign was only found in the infant where this had already been diagnosed prenatally.
In our experience, intestinal gas frequently prevents visualisation of the superior mesenteric artery sufficiently to detect/rule out the whirl pool sign during abdominal ultrasound in infants with acute abdomen.
The common practice of abdominal massage has also been associated with volvulus.1 2 This was also here reported for 12 of 15 infants, but its relevance to volvulus remains unclear given our uncontrolled study design.
In seven infants, surgical intervention was fast enough to avoid any bowel resections; at least 4 of these infants were eventually discharged home. One infant had no resection performed despite more than 90% of the intestine being ischaemic. This infant died of multiorgan failure prior to an intended second-look operation. Of the 18 infants who did have a bowel resection, at least 13 were discharged home, although four developed SBS (15% of all verified VWM cases), with three of the four continuing to grow normally on full enteral nutrition. In general, the development of an SBS is rare (0.7%), as shown in a cohort of 12316 VLBW infants, mainly caused by NEC.11
In the case study of Drewett and Burge, none of the infants develop an SBS. The total parenteral nutrition lasted a median of 12.5 days, and all affected infants could be discharged on full enteral feeding.8 The oral feeding in the study by Billiemaz et al 1 could be started at a median of 11 postoperative days; 14.3% developed an SBS and 71% were discharged with full oral feeding.
Information on the further course is sparse in other case reports. Four case studies1 2 6 9 mentioned data on the proportion of SBS with a median value calculated from these reports of 12.7%.
Mortality (10%) was lower than reported in all previous case series,1 3 5 6 8 12 but this proportion might underestimate actual mortality rate as there was no information on survival for six infants, and therefore a robust conclusion regarding mortality cannot be made.
Limitations
Our study is hampered by incomplete questionnaire data despite several attempts to improve response rates and by its uncontrolled nature. Nonetheless, it constitutes the first attempt to generate population-based data on this rare and dramatic event. Data from other surveillance units that ideally include controls are urgently awaited.
Conclusion
VWM is a life-threatening disease, resulting in SBS in a substantial proportion of survivors. Differentiation from other acute abdominal conditions in preterm neonates remains challenging. Unfortunately, our survey did not identify any specific diagnostic sign. We also failed to identify clinical or imaging findings that would make VWM highly unlikely. Therefore, an immediate explorative laparotomy may be considered in any preterm infant with signs of an acute abdomen and deterioration of the clinical situation to prevent irreversible bowel ischaemia.
Acknowledgments
We would like to thank the German Surveillance Unit for Rare Pediatric Diseases (Erhebungseinheit für seltene pädatrische Erkrankungen in Deutschland Study Group (ESPED)) and especially Beate Heinrich for their support and all colleagues from the participating hospitals for reporting their cases.
Footnotes
Contributors CM designed the study and the questionnaire, supervised data analyses and reviewed and revised the manuscript making important intellectual contributions. ARF was study coordinator, contributed to study and questionnaire design, supervised data analyses and reviewed and revised the manuscript making important intellectual contributions. YY was responsible for the data collection and recording and reviewing and revising the questionnaires. H-JK reviewed and revised the manuscript for important intellectual contributions. CFP supervised the study as the head of department and reviewed and revised the manuscript for important intellectual contributions. All authors read and approved the final manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Disclaimer The authors have indicated the statements in the submitted article are their own and not an official position of the institution.
Competing interests None declared.
Ethics approval The study protocol, including a parental consent waiver, was approved by the ethics committee of Tübingen University Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement There is no additional unpublished data available.