Safety of Pregnancy in Ventriculoperitoneal Shunt Dependent Women: Meta-analysis and Systematic Review of the Literature
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.288995
Source of Support: None, Conflict of Interest: None
Keywords: Hydrocephalus, pregnancy, shunt, systematic review, ventriculoperitoneal, women
Hydrocephalus (HCP) is a neurosurgical disorder which is defined as the presence of an abnormal collection of cerebrospinal fluid (CSF) inside the brain ventricles. This disorder affects both pediatric and adult population with an estimated prevalence of 0.9–1.2/1.000. HCP is further classified as communicating and non-communicating (obstructive) HCP. The most common etiology of the HCP with an incidence of about 0.4–1.0 in 1,000 births is congenital causes including Chiari Type 1 and 2, primary and secondary aqueduct stenosis, Dandy Walker malformation and X-linked inherited disorders., The other causes of the HCP are acquired. The HCP is more commonly diagnosed in children than in adults which is usually related to anatomical malformation within the brain. However, the causes vary in adults, examples include head injuries, prior operations, tumors, aqueduct stenosis, subarachnoid hemorrhage, meningitis, and idiopathic.
Various types of shunts were invented during the period of 1898 and 1925 such as lumboperitoneal, and VP, -venous, -pleural, and –ureteral shunts. Due to high failure rate because of insufficient implant materials in most cases, the introduction of silicon CSF shunts in the 1960s have decreased the mortality of patients with HCP with more women reaching reproductive age. VP shunts are the most common and effective procedure for the treatment of HCP.
A recent search of the current literature demonstrates that a broad range of interdisciplinary professionals, including neurosurgeons, obstetricians, neurologists, perinatologists, genetics counselors, and other healthcare professionals, are increasingly interested in this area of study. A variety of maternal and fetal complications (obstruction, displacement, infection, and increased risk of congenital hydrocephalus in their offspring associated with VP shunts) are described in the literature. Aspiration of CSF was found to be the best treatment of choice for VP shunt malfunction during pregnancy. Whereas the replacement of VP shunt with VA shunt may be considered for patients requiring a reinsertion of the VP shunt. There are no general recommendations for delivery with these patients so the decision depends on the recommendation of the neurosurgeon for each patient. Almost half of the cases of pregnancies with shunts have been reported to suffer from increased intracranial pressure. Signs and symptoms of shunts obstruction can involve severe headache, nausea, vomiting, visual disturbance, ataxia confusion, and papilledema., Danijel Bursac claimed that the best way to help a patient with a VP shunt in giving a birth is spontaneous vaginal delivery, with epidural analgesia, mediolateral episiotomy, and vacuum extraction, to make the second stage of labor very short. However, there are no previously published controlled studies about the management of pregnancy and labor in women with VP shunts. To date, there is no a comprehensive review on pregnancy safety of women with VP shunt despite the increasing number of these women reaching their reproductive age. Therefore, the aim of this study systematically review and summarize the available evidence on the safety pregnant women with VP shunt in order to establish the first clinical recommendations on pregnancy management in these women.
Data sources and search strategies
The PRISMA Statement was used to guide the conduct and reporting of this review 1. A literature search was done using three electronic databases (MEDLINE via PubMed, EMBASE, the Cochrane Library) from 1950 to 2019. Additionally, we searched the references at Web of Science, Google Scholar, and ClinicalTrials.gov. A variety of words “ventriculoperitoneal shunt,” “dependent,” “hydrocephalus,” “safety,” “pregnancy,” and “delivery” were used as Mesh terms and keywords.
Inclusion and exclusion criteria
Studies were included if they met the following inclusion criteria: (i) were case reports and case series, (ii) reported cases of pregnant women with ventriculoperitoneal (VP) shunt, (iii) were published in English language. There was no age restriction and articles from around the world were searched. However, the cases of pregnant women with a shunt other than VP were excluded. Two reviewers independently evaluated the titles and abstracts according to the inclusion and exclusion criteria. For each potentially eligible study, two reviewers assessed the full-text. In cases of disagreement, a decision was made in consultation with the third reviewer.
Data extraction and assessment
A predesigned data extraction form was used to collect relevant information. In particular, we extracted the following data: the name of the first author, year of publication, age of the patients, reason of shunting, time of shunting, delivery mode and gestation age, type of anesthesia, complication (antepartum, postpartum), and neonatal outcome.
Given that there are no validated tools to assess the methodological quality of case reports and case series, we applied the quality assessment tool suggested by Murad et al. and utilized in previous publications.,, The items were derived from the Newcastle-Ottawa Scale (NOS). The items that relate to comparability and adjustment were excluded while the items focused on selection, representativeness of cases, and ascertainment of outcome and exposure were retained. Therefore, 5 criteria in the form of questions with yes/no response were used to judge the quality of studies included in the review. Therefore, the questions answered for each case report were: (i) Did the patient (s) represent the whole case (s) of the medical center? (ii) Was the diagnosis correctly made? (iii) Were other important diagnoses excluded? (iv) Were all important data cited in the report? and (v) Was the outcome correctly ascertained? Accordingly, the quality of evidence was considered to be either low (≤ 3 criteria fulfilled), moderate (4 criteria fulfilled) or high (all criteria fulfilled).
Overall 58 references were identified using the search strategy. After removing 21 duplicates and after the initial screening based on titles and abstracts, the full texts of 37 articles were retrieved and evaluated in details. After full text assessment, 11 articles were excluded based on pre-defined exclusion criteria leaving 26 articles to be included in this systematic review [Figure 1]. All publications were in English, 10 were from Europe, 6 from Asia, 9 from North America and 1 from Africa. Nineteen publications were case reports while 7 were case series comprising in total 38 cases.
Characteristics of cases included in the current review
Among the 38 cases of pregnant VP shunt-dependent women, median age was 25.5 years and shunting duration was 15.5 years with 11 women being shunted at birth or soon after [Table 1]. Congenital diseases associated with HCP were the most common reason for shunting, present in 63.2% of women. The antepartum complications were reported in 19 (50%) of cases with the symptoms of increased ICP being the most commonly reported (73.7%). In the majority of cases the complications were resolved with CSF aspiration (26.3%). Shunt removal and endoscopic third ventriculostomy (ETV) insertion was indicated in 15.8% of women with complications, while in 21.1% of cases the intervention was not needed [Table 1]. Detailed information on cases characteristics can be found in Online [Table 1].
Safety of pregnancy in ventriculoperitoneal shunt dependent women
Among the 38 cases, a half of the women did not present with symptoms during the pregnancy. Median age of these women was 21 and median shunting duration was 19. The majority of women with no symptoms had a congenital disease as an indication for shunting (57.9%). Eight women (42.1%) had spontaneous vaginal delivery, 4 had assisted vaginal delivery, while 7 women underwent cesarian section. In 7 women, postpartum antibiotic prophylaxis was initiated. All newborns, besides a single one, were born healthy or with no complications; one new-born presented with mild respiratory distress which was successfully resolved. Postpartum complications were reported in three women only: shunt malfunction , and shunt infection  were reported in two and one women respectively. In two cases, complications were successfully resolved with shunt replacement  and in one case with CSF aspiration and distal catheter cut-ff. [Table 2].
Raised ICP symptoms were the most common symptom, present in 73.68% of women, with antepartum complications. The median age of women was 26.5 years and median shunting duration was 15 years. Similarly, to women with no complication the most common indication for shunting were congenital diseases (71.4%). The majority of women delivered babies via cesarian section (n = 8), while five women had spontaneous vaginal and one women had induced vaginal delivery. Postpartum complications were reported in two women only: (i) headache and diplopia resolved with valve pressure change  and (ii) seizure and loss of conciseness which led to VP shunt replacement by VA shunt. All newborns were born healthy or with no symptoms. However, one intrauterine fetal death has occurred in a known case of tuberous sclerosis mother who presented with status epilepticus during the pregnancy. In this woman second VP shunt was inserted and virginal delivery was inducted at gestation age of 20 weeks. Initially, she was treated with antiepileptic medications, intubated and admitted to intensive care unit. The patient was not known to be pregnant but a routine urine pregnancy test came as positive. Then, an ultrasound showed a single live fetus of about 20 weeks. A CT scan confirmed the diagnosis of blocked VP shunt and was replaced by another shunt in theatre. After 5 days of admission, no fetal heart sounds could be detected. Then, she was administered mifepristone. Three days after, a vaginal misoprostol was administered in the delivery word and a 450 g dead infant was delivered. [Table 2]
Postpartum complications in ventriculoperitoneal shunt-dependent women
Postpartum complications were reported in only 5 (13.2%) among 38 women included in the current review. Shunt malfunction and shunt infection were reported in four and one women respectively. Women's age varied from 19 to 34 years, and shunting duration varied from 3 to 34 years. Among these women only two women had complications during pregnancy., Both presented with headaches and one had memory disturbances and altered mental status. All complications were successfully resolved and 3 newborns were born healthy and in two cases the new born status was not reported [Table 3].
Generalizability of the results
The majority of cases included in current review were evaluated to have high quality (n = 24), while 9 to have moderate and 5 to have low methodological quality [Table 4]. Given the high and moderate quality of most included studies, the location of patients in 14 different countries, and the scarce reports of this phenotype, we believe that our results could be applied to all pregnant women dependent of VP shunt. However, we could not exclude a selection bias favoring the reporting of more severe cases.
Variable perspectives regarding VP shunt-related complications were delegated in the literature review of pregnant women. It sheds the light on the importance of a multidisciplinary approach to increase maternal and fetal safety by emphasizing the use of variable precautionary measures. It is also important to emphasis the research critical to advancing the knowledge base from which we draw information in order to make the most effective clinical management decisions. There is some controversy surrounding the treatment of shunting for pregnant women. Special management is required for Women with VP shunts during the period of pregnancy with a team of anesthesiologists, neurosurgeons, and obstetricians. Interdisciplinary management has to be performed during the preconception, antenatal, and postnatal periods as well as at the time of delivery.,, [Table 1] briefed the 38 reported pregnancies with VP shunt. Congenital causes was found to be the common reason for shunting including: primary aqueduct stenosis with HCP, meningomylocele with HCP, Arnold Chiari malformation with syringomyelia and Dandy Walker malformation.,,,,,,,,,,,,, Almost half of pregnancies had no antepartum complications. Haeussler et al. reported three cases in their series with VP shunts had no shunt-related complications during these pregnancies. Shunted women with a congenital disease seemed to have the least symptoms. The other half of the cases have been noticed to have antepartum complications with increased ICP being the most commonly reported. The malfunction of shunts in pregnant women remains a challenging medical condition for both obstetricians and neurosurgeons.
No standard procedure has been established yet for shunt malfunctions during pregnancy. However, low rates of shunt malfunctions and revisions were found in pregnant women. In our review, the complications were resolved in majority of cases with CSF aspiration (26.3%). Flushing device for the blocked shunt system was discussed by Sasayama et al. In 21.1% of cases with complications, neurosurgical interventions were not required. Whereas shunt removal and ETV insertion was indicated in 15.8% of women with complications. For the management of VP shunt malfunction during pregnancy Murakami et al. recommended ventriculoatrial (VA) shunt replacement as a treatment option. Riffaud et al. used ETV as a treatment modality in newly diagnosed obstructive HCP and malfunction of a preexisting shunt. The use of an anti-siphon device shunt was studied before in the treatment of HCP in pediatrics and normal pressure HCP patients, but some work on the use of ASDs for treating pregnancy HCP might be considered as an alternative to standard VP shunt device.
During pregnancy, the expanding gravid uterus may cause an increase in intraperitoneal pressure. This pressure increase distal to the shunt system causes a decrease in the pressure gradient between the ventricle and peritoneal cavity, which consequently results in decreased shunt flow. The flow of the shunt must be kept optimum, neither too small nor too large, otherwise the pressure gradient decreases and results in a decreased shunt flow. The repeated VP shunt malfunctions in this present review seems to derive from a combination of factors; which are the narrow optimal range of the shunting pressure, the patient's high shunt dependency due to congenital causes, and the pressure increase in the intraperitoneal cavity which causes a reverse pressure gradient in the shunt system.
There is still uncertainty in regards to mode of delivery, analgesia, and anesthesia of women with VP shunt. Some studies recommend Cesarean section in patients who are likely to go into preterm delivery. Freo et al. reported that a rapidly deteriorating level of consciousness necessitates urgent and simultaneous cesarean delivery and neurosurgical intervention.
In our review, eight women (42.1%) had spontaneous vaginal delivery, 4 had assisted vaginal delivery, while 7 women underwent cesarian section. Some anesthesiologists preferred general anesthesia over spinal anesthesia because general anesthesia is relatively safe, could decrease cranial pressure and it eliminates the possibility of VP shunt infection that may arise with spinal anesthesia. Vaginal delivery is preferable in women with VP shunts, and assisted delivery or Cesarian section should be reserved for obstetrical indication only.,,,
There are some articles debating on prophylactic antibiotic use both in vaginal delivery and Cesarian section for patients with VP shunts., In seven women, postpartum antibiotic prophylaxis was initiated. Different types of antibiotics were used as single agent or combination including: ampicillin, gentamicin, vancomycin, meropenem, and cefazolin. Considering the potential risks of septicemia seeding of the VP shunts, prophylactic antibiotics seem justified, though not of proven benefits.,,,, All newborns, besides a single one, were born healthy or with no complications; one new-born presented with mild respiratory distress which was successfully resolved. There was one fetal demise occurred in a woman who was diagnosed with tuberous sclerosis and presented with status epilepticus during the pregnancy. She required intubation in the intensive care unit and multiple shunt revisions. Postpartum complications were reported in three women only: shunt malfunction , and shunt infection (S2) were reported in two and one women respectively. In two cases, complications were successfully resolved with shunt replacement  and in one case with CSF aspiration and distal catheter cutoff. Nikolov et al. studied 12 pregnant cases (9 with VP shunts and 3 with VA shunts) and concluded that the presence of a cerebral shunt does not affect the pregnancy.
The management of VP shunts during the pregnancy and post-partum period, which is considered as the first step toward insuring the best pregnancy outcome for both mothers and the fetus requires multidisciplinary teams including neurosurgeons, neurologists, obstetricians, perinatologists, genetics counselors and other healthcare professionals. [34,35] The current review presents a cutting-edge summary of pregnancy management in women with VP shunt. However, the recommendations from this paper shall be interpreted with caution. The quality of case reports included in this review was in general high or moderate, however, there are no previously published controlled studies about the management of pregnancy and labor in women with VP shunts which would improve the quality of evidence. The most important clinical recommendations based on the current review are:
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3], [Table 4]