A parturient with varicella (chickenpox) presented for an elective Caesarean section and spinal anaesthesia was employed for surgery. A review of the literature is presented and the anaesthetic issues are discussed.
Varicella zoster virus (VZV), a DNA virus belonging to the herpes virus family, causes a primary contagious and usually benign illness commonly known as varicella or chickenpox. The virus can also lie dormant in the dorsal root ganglia and may be reactivated and result in localised cutaneous eruptions called ‘herpes zoster’ (‘shingles’). The lifetime risk of VZV infection is about 95%. About 5–10% of pregnant women lack antibodies to VZV and may acquire varicella during pregnancy . Although only about 2% of all cases occur in adulthood, they account for 25% of all VZV-related deaths . Varicella pneumonia is 25 times more common in adults . The consequences of primary VZV in pregnancy for the mother and for the fetus vary with the period of gestation. The risk of adverse effects for the mother is greatest in the third trimester, whereas for the foetus the risk is greatest in the first and second trimesters .
In this case report, we describe a case of an obstetric patient recovering from varicella (chickenpox) that presented for an elective Caesarean section, and discuss the anaesthetic issues related to varicella in pregnancy.
A 41-year-old gravida 3 para 2 woman, who had experienced two previous uneventful lower segment Caesarean sections (LSCS) under epidural anaesthesia, presented for an elective LSCS at 39 weeks' gestation. Her obstetrician noted active varicella lesions on her trunk. She was febrile (38.1 °C) and had a productive cough. The patient revealed that the lesions had first erupted about 2 days prior to the scheduled visit. She had not been previously immunised and could not recall any exposure to varicella. There were no other significant physical findings. On the advice of the medical virologist, the patient was treated with varicella immunoglobulin and the procedure was deferred for about a week.
She re-presented for the elective LSCS 9 days after treatment with varicella immunoglobulin (11 days following the first appearance of the varicella lesions). At this stage, she was afebrile and the skin lesions were dry and crusted. The cough had improved. The patient was keen to be awake for the procedure. The potential risks of a neuraxial block in the presence of a viral infection were discussed and a spinal anaesthetic was chosen. A spinal anaesthetic, using a 24-gauge Sprotte needle and 2.1 ml 0.5% bupivacaine with 20 μg fentanyl, was performed at the L2/L3 level, where there were no varicella lesions on the skin. The surgery proceeded uneventfully with a live male infant (3600 g) delivered with Apgar scores of 9 at 2 min and 5 min after delivery. The infant had grunting respiration that improved after oro-pharyngeal suction and oxygen therapy, and 15–20 varicella lesions on the face. The infant received a course of intravenous acyclovir. The mother had an uneventful postoperative course.
The use of regional anaesthesia in obstetric patients with viral infections such as active herpes infections and human immunodeficiency virus (HIV) is controversial . The main concern of performing a spinal or epidural anaesthetic is the possibility of introducing the virus into the central nervous system during the placement of the block, resulting in meningitis or encephalitis, especially during the primary infection when viraemia is present. Further, if postoperative headache or neurological sequelae occur, it may be difficult to exclude the neurological complications of the neuraxial blockade from those arising from the disease. In parturients with HIV infection, there is no evidence of spread of HIV to the central nervous system through neuraxial blockade . Neuraxial blockade is not recommended in primary herpes infections because these patients have viraemia . However, it is safe in recurrent herpes infection because viraemia is absent .
Varicella (chickenpox) is a highly contagious disease with an infection rate of at least 90% among seronegative or susceptible individuals. Children are the most commonly affected and account for more than 50% of cases. The VZV is transmitted by direct contact or via aerosolised respiratory droplets. The virus replicates in the nasopharynx with subsequent episodes of viraemia. The incubation period of varicella is between 10 and 21 days. Patients are infectious approximately 48 h prior to the onset of the rash, during the period of vesicle formation (usually 4–5 days) and until all vesicles have crusted. Clinically, varicella presents as a rash with a low-grade fever and malaise. Prodromal symptoms may develop 1–2 days before the appearance of the rash. The skin lesions evolve from a maculopapular rash to vesicles. The lesions appear in the trunk and face and rapidly involve other areas of the body. Lesions may be found on the mucosa of the pharynx or vagina. The lesions have an erythematous base with a diameter of 5–10 mm. Secondary bacterial superinfection caused by Streptococcus pyogenes or Staphylococcus aureus is the most common complication of varicella. The most common extra-cutaneous site of involvement is the central nervous system. Acute cerebellar ataxia, encephalitis aseptic meningitis and Guillain–Barré syndrome have been reported in VZV infections .
There are several issues in the anaesthetic management of this patient. A regional anaesthetic technique was used for two reasons, the patient's desire to be awake during the procedure, and the potential risk of varicella pneumonia. Varicella pneumonia is the most serious complication following varicella, occurring more commonly in adults than in children. It appears 3–5 days into the illness. Before the availability of antiviral therapy, 65% of parturients with VZV infections had varicella pneumonia (usually occurring in the third trimester). With antiviral therapy this prevalence has decreased to 3.6–9%, comparable to the non-pregnant population [9–11]. The clinical manifestations of varicella pneumonia usually coincide with the pre-existence of the characteristic rash and include fever, dyspnoea, cough, pleuritic pain and haemoptysis. The characteristic chest radiographic findings are bilateral nodular infiltrates and interstitial pneumonitis, usually detected at peak severity of the rash. Gambling & Douglas recommend regional anaesthesia as the preferred choice in a parturient with an acute VZV because of the high risk of varicella pneumonia . However, Camann & Tuomala suggest that regional anaesthesia should be avoided for 2 weeks after the onset of symptoms of varicella because of viraemia . However, these recommendations are not based on randomised controlled studies or evidenced-based medicine because varicella in pregnant women is not common. Tissue coring with hollow spinal needles resulting in the spread of epithelial cells into the epidural and intrathecal spaces has been reported during epidural analgesia and lumbar puncture [14, 15]. We suggest that the pencil point needle used in this patient may reduce the potential risk of introducing infective viral material into the central nervous system. Further, active (from the VZV infection) and passive (via varicella immunoglobulin) immunological protection acquired by our patient would confer additional protection. It is important to stress that the spinal anaesthesia should be performed at a level where no skin lesions are present. There is a theoretical advantage in using epidural rather than spinal anaesthesia because an epidural needle does not breach the dura and theoretically would lessen the risk of introducing the virus into the central nervous system. However, this theoretical advantage may be lost because of the potential to introduce a larger viral load by epidural needles that are hollow and larger in size.
What would have been the optimal anaesthetic management if this patient presented for an emergency Caesarean section? There are no evidence-based guidelines available. We suggest that the respective benefits and risks of general anaesthesia and neuraxial blockade have to be considered. General anaesthesia would be the anaesthetic of choice if there were active or infected lesions on the skin at the site for placement of a spinal or epidural block. No data are available on the relative risks of varicella pneumonia with general anaesthesia or the spread of virus with a regional block in this scenario. Harger and his colleagues, in a case–control analysis, compared 18 pregnant women with varicella pneumonia with 72 matched control pregnant women with varicella without pneumonia . In this multicentre study 347 pregnant women who contracted primary varicella infection during any trimester were enrolled. Multivariate logistic regression analysis showed that two factors significantly correlated with an increased chance of varicella pneumonia: smoking during the affected pregnancy (odds ratio = 5.1; 95% CI, 1.6–16.7) and > 100 skin lesions (odds ratio = 15.9; 95% CI, 1.9–130.2). The 18 women with varicella pneumonia all received intravenous acyclovir therapy and Caesarean delivery was undertaken in six patients. Five Caesarean deliveries were performed for obstetric reasons and one for acute maternal respiratory failure. This patient developed respiratory failure at 37 weeks' gestation that required mechanical ventilation. A Caesarean delivery was performed as her condition deteriorated. The mother recovered without sequelae after 7 days of acyclovir therapy. The anaesthetic management was not described in the study, but only one patient was intubated and received mechanical ventilation for 6 weeks because of bacterial superinfection. There were no maternal deaths in this study, which concluded that the lack of maternal mortality in this case series of women with varicella pneumonia was the result of early diagnosis and vigorous antiviral therapy. The safety of oral or intravenous acyclovir during pregnancy and its benefits were uncertain in studies prior to 1996. If general anaesthesia is used in a pregnant woman with active varicella, we suggest that early diagnosis of pneumonia and vigorous antiviral therapy may be useful in reducing the risks of varicella pneumonia following general anaesthesia. If regional (neuraxial) anaesthesia is used, we suggest that meticulous neurological examination and monitoring of the patient should be performed peri-operatively and that early antiviral therapy should be considered.
Exposure of medical personnel to an infectious patient should be avoided or minimised. Although previous exposure or infection with VZV is regarded as conferring immunity, second varicella infections can occur and are more common than previously thought . Prophylactic administration of zoster immunoglobulin (ZIG) can prevent or reduce the severity of varicella . For maximal effect, ZIG should be administered within 72 h of exposure, although it may provide some benefit 96 h after exposure for immunocompromised subjects . This patient received ZIG (625 U) approximately 48–72 h after the onset of symptoms with the aim of attenuating the infection and also ensuring adequate transplacental transfer of the immunoglobulin to the foetus . When varicella occurs in the first trimester of pregnancy, close ultrasound monitoring for the development of foetal abnormalities after maternal varicella is recommended .
Foetal infection with VZV occurs in 10–15% of cases of varicella in pregnant women but is usually transient and asymptomatic. Shingles is the most common clinical manifestation in the first year of life. Approximately 2–3% of infants of women who have chickenpox in the first half of pregnancy develop foetal varicella syndrome, characterised by dermatomal skin lesions or scarring, unilateral limb hypoplasia, visceral, eye and neurological lesions . The timing of maternal infection in relation to delivery determines the risk to the infant. Maternal infections occurring between 4 and 7 days before delivery or 2 days after delivery is associated with an increased risk of severe neonatal varicella in the infant caused by the transplacental spread of the virus. Maternal infection with onset more than 7 days before delivery is associated with less severe neonatal varicella because of adequate transplacental passage of VZV antibodies to protect the foetus. Infants born to mothers who acquired varicella infection should be given ZIG as soon as possible after birth. In this case report, the infant, who had the characteristic facial rash, had respiratory symptoms and was treated with acyclovir. However, routine acyclovir prophylaxis in addition to the ZIG is not recommended in the neonate because of lack of evidence . To prevent severe neonatal chicken pox, passive immunisation is indicated. If varicella occurs, acyclovir treatment must be administered promptly .
Maternal varicella pneumonia is a serious complication of VZV infection. The incidence of varicella pneumonia is reported as 1 in 10 000 pregnancies . The frequency of varicella in all adults was estimated to be 1.2–6 per 100 000 admissions in Stockholm and Scotland [24, 25]. Smoking is a risk factor for varicella pneumonia in adults [16, 26]. Women in the third trimester are at a greater risk for developing VZV pneumonia . Other risk factors for VZV pneumonia in parturients are more skin lesions (> 100) and the presence of pharyngeal varicella lesions . Intravenous acyclovir should be given for varicella pneumonitis at any stage of pregnancy [16, 27, 28]. The recommended dose of acyclovir in the treatment of varicella pneumonia is 7–10 mg.kg−1 intravenously three times daily for 7 days. Extracorporeal membrane oxygenation may have a role in managing severe varicella pneumonia [29, 30].
A live attenuated vaccine for the prevention of varicella has been available since the mid-1980s. Varicella virus vaccination provides > 70% protection against infection and prevents severe disease in 90% of vaccinated subjects for 7–10 years. Susceptible women of childbearing age should be vaccinated to reduce risks of serious morbidity associated with adult infection and also to reduce vertical infection to children. Women should be advised to avoid pregnancy for 1 month after vaccination. Vaccination is not recommended during pregnancy . Although there are no cases of congenital varicella syndrome associated with inadvertent vaccination during pregnancy, the number of cases evaluated is too limited to make any conclusions concerning the safety and risks of vaccination during pregnancy.
The potential risks of varicella infection among adults are a concern and challenge to the clinician, and have important implications for susceptible pregnant women. Prompt recognition, evaluation and treatment are important in preventing serious maternal complications. Varicella pneumonia must be excluded in pregnant women with signs of primary varicella and intravenous acyclovir is warranted when evidence of varicella pneumonia/pneumonitis are present. Primary prevention with vaccination of seronegative women of childbearing age, and secondary prevention with postexposure prophylaxis with ZIG are key strategies among pregnant women and non-pregnant women of childbearing age.