Spinal anaesthesia for Caesarean section in a patient with systemic sclerosis


A. R. Bailey , Specialist Registrar in Anaesthesia, Norfolk and Norwich Health Care NHS Trust, Brunswick Road, Norwich NR1 3SR, UK


We describe the management of a diabetic primigravid woman with systemic sclerosis and thrombocytopaenia who required Caesarean section for pre-eclampsia. This was performed successfully under spinal anaesthesia.

Systemic sclerosis (scleroderma) is a multisystem disease of unknown aetiology, with an annual incidence of 2.3–10 per million in the adult population and a predilection for females (female : male ratio of 10 : 1 in the reproductive age group 15–44 years) [1]. It is characterised by the overproduction and growth of collagen, widespread vascular damage and the development of microvascular obliteration. While the exact effect of pregnancy on the progression of this disease and the effect of the disease on pregnancy itself are unclear, it is apparent that the disease has certain implications for the conduct of anaesthesia. Although the use of general anaesthesia for Caesarean section [2[3]–4] and epidural analgesia for labour [5] have been reported in women with this disease, regional anaesthesia for Caesarean section has not. We describe a case of a pregnant woman with systemic sclerosis who required anaesthesia for Caesarean section.

Case history

A 30-year-old nulliparous woman was admitted to hospital from the antenatal clinic at 32 weeks' gestation with signs and symptoms of pre-eclampsia, notably headache, diastolic blood pressure of 95 mmHg, 2+ proteinuria and peripheral pitting oedema. In addition to being a well-controlled insulin-dependant diabetic, she had a 14-year history of systemic sclerosis with Raynaud's phenomenon, nonpitting oedema and ‘hidebound’, thickened, sclerotic, skin, which were present prior to this pregnancy and had continued unchanged throughout. Several years previously she was also noted to be thrombocytopaenic, with a platelet count of 90 × 109.l−1, the aetiology of which remained uncertain.

On admission, her haemoglobin was 10.2 g.dl−1, platelet count 104 × 109.l−1 and the clotting screen was normal. Cardiotocograph recording was normal. Over the subsequent 4 days, her symptoms persisted and her blood pressure remained elevated. In addition, her platelet count continued to decrease, serum urate remained raised and she had 3.12 g proteinuria in 24 h. Her blood glucose varied between 7 and 11 mmol.l−1. A glucose–insulin–potassium regimen was commenced for diabetic control, two doses of intramuscular betamethasone 12 mg were administered 24 h apart to enhance fetal lung maturity and Caesarean section was scheduled for the next elective list. The rheumatology team was consulted, and felt that the hypertension and proteinuria represented pre-eclampsia rather than renal scleroderma.

At pre-operative anaesthetic assessment it was noted that the patient had limited mouth opening, with a class 4 Mallampati score. Blood pressure was 170/100 mmHg, 4+ proteinuria was present and the platelet count was 61 × 109.l−1. Coagulation screen was normal and the bleeding time was 3 min. A two-dimensional echocardiograph revealed a mildly hypertrophied, but well-contracting, left ventricle, while Doppler demonstrated trivial pulmonary and mitral regurgitation.

In view of the potential for difficult intubation and increased risk of regurgitation in patients with systemic sclerosis, it was decided that regional anaesthesia would be preferable. A regional technique permitting gradual extension of anaesthesia would have been the most suitable approach. However, in the presence of a worsening thrombocytopaenia it was considered that epidural catheterisation was contraindicated. Therefore the patient was prepared for Caesarean section under spinal anaesthesia. The procedure was explained and cimetidine 200 mg was administered 2 h pre-operatively.

On arrival in the anaesthetic room, a 16 G venous cannula was inserted into the right hand, an intravenous infusion established, oxygen at 4 l.min−1 was administered and routine monitoring applied (noninvasive blood pressure, ECG and pulse oximetry). Spinal anaesthesia was then performed at the L2/3 interspace, with the patient sitting. A dose of 4.75 ml of 0.25% plain bupivacaine with 0.25 mg diamorphine was injected via a 25 G Sprotte needle, a technique routinely used by one of the authors (SR). The patient was then laid supine with left lateral tilt. A bilateral sensory block to cold was achieved to T2 with complete lower limb motor block and the patient was transferred to the prewarmed operating theatre. Prior to institution of spinal anaesthesia, the patient's blood pressure was 160/90 mmHg but this decreased to a systolic pressure of 90 mmHg shortly after establishment of the block, despite an ephedrine infusion being commenced at 15 mg.h−1 immediately after performing spinal anaesthesia. A further 42 mg of ephedrine, administered incrementally, was required during the procedure to maintain the systolic blood pressure between 120 and 140 mmHg. Caesarean section was performed uneventfully with delivery of a baby girl with Apgar scores of 6 and 10 at 1 and 5 min, respectively. Oxytocin 5 units was administered intravenously after delivery and a further 5 units were given 5 min later. A total of 2 l of fluid were infused, with blood loss amounting to ≈ 600 ml. Peroperatively, all fluids were warmed and all limbs wrapped in ‘Gamgee’ for both warmth and protection. The pulse oximeter probe was changed between fingers to prevent causing ischaemic damage to the patient's already poorly perfused digits.

Full sensation returned 3.5 h after institution of spinal anaesthesia. Postoperatively, the patient was nursed and monitored on the delivery suite for 24 h. Analgesia was provided by intravenous diamorphine using a patient-controlled analgesia system. Over the subsequent 4 days, the hypertension and proteinuria resolved and she resumed control of her diabetic management with subcutaneous insulin. She was discharged home 5 days following delivery.


Scleroderma is a heterogeneous disorder, often presenting with Raynaud's phenomenon and ranging from the more common mild cutaneous form, with minimal or late-onset organ involvement (limited cutaneous systemic sclerosis) to an aggressive, life-threatening, diffuse form (diffuse cutaneous systemic sclerosis) [6]. While the 10-year survival rate is 50% with limited cutaneous systemic sclerosis, it is 25% with the diffuse disease [7]. Genetic and environmental factors are involved in the aetiology, with a high incidence of anticentromere antibodies in the limited cutaneous form of the disease and an association with the major histocompatibility complex [8]. The immune system is involved in initiating the fibrotic process, while impaired regulation of collagen gene expression probably underlies the persistence of fibrosis [9].

Although systemic sclerosis is presently incurable, immunosuppressive agents (including cyclosporin, methotrexate and photophoresis) are used to modify the disease progress. Interferon (IFN)-α, with its potential antifibrotic effect, may also have a role in management, while nifedipine and nonsteroidal anti-inflammatory drugs may be used for symptomatic relief [9].

The clinical effects of excess fibrosis and vascular damage seen in systemic sclerosis have important implications for the anaesthetist. Venous access may be impaired by the thickened skin and chronic vasoconstriction. Flexion contractures and vasoconstriction impair indirect blood-pressure measurement, while direct arterial cannulation and blood-pressure measurement may be associated with vasospasm and distal necrosis [10]. Ultrasonic measurement of blood pressure has proved useful [5]. Skin contractures and microstomia make tracheal intubation difficult, nasal and oral telangiectasiae bleed if traumatised, while oesophageal dysmotility and sphincter incompetence can lead to aspiration. Pulmonary fibrosis may occur and pulmonary hypertension may arise especially in the late stages of the limited cutaneous form. Cardiac involvement manifests as myocardial fibrosis, pericarditis, arrhythmias and conduction defects.

Renal disease is a major cause of mortality in systemic sclerosis, especially in pregnant patients with the disease [11]. Histologically there is an intimal thickening of interlobar arteries. Several presentations have been reported, including chronic proteinuria and uraemia, hypertension (occasionally malignant), microangiopathic haemolytic anaemia and rapidly progressive renal failure [12]. Angiotensin-converting enzyme inhibitors are the first line of treatment to control the hypertension of renal systemic sclerosis and prevent further renal deterioration, despite concerns about their safety in pregnant women [13]. During the second half of pregnancy, differentiation between renal scleroderma and pre-eclampsia may be difficult. Seizures, elevated transaminases or urate may assist in the diagnosis although, if the situation remains unclear, measurement of plasma renin may prove useful. In pre-eclampsia, the serum renin is low to normal while, in the renal crisis of systemic sclerosis, plasma renin is elevated as a result of renocortical ischaemia [14].

The effect of systemic sclerosis on mother and fetus during pregnancy and the effect of pregnancy on disease progression is uncertain. The disease onset is often beyond the reproductive years, the natural course involves exacerbations and remissions and the literature contains case reports and series which are undoubtedly subject to reporting bias. In reviewing the available literature, Black [1] concludes that the rates of infertility, spontaneous abortion, prematurity and intrauterine growth retardation may be higher in women with systemic sclerosis, while a 15% incidence of maternal death has been reported, mainly resulting from renal and cardiopulmonary complications [14].

Both pregnancy and systemic sclerosis are associated with risks of difficult intubation and aspiration of gastric contents. Nevertheless there are occasions when regional anaesthesia is contraindicated and general anaesthesia is the only option for Caesarean section. Under these circumstances, awake tracheal intubation should be considered for the patient with systemic sclerosis, especially when mouth opening is limited, as spontaneous respiration will be maintained should difficulties arise and the cough reflex is preserved if regurgitation occurs [2, 3].

When possible, a regional technique is preferable for the patient with systemic sclerosis, avoiding the risks of aspiration and failed intubation. Nevertheless, regional anaesthesia should be performed cautiously to avoid total spinal or inadequate anaesthesia requiring emergency tracheal intubation. In addition, Eisele and Reitan [15] have suggested that smaller doses of local anaesthetic are required, as systemic sclerosis patients exhibit prolonged sensory and motor blockade. Insertion of a catheter into the epidural or intrathecal space allows the gradual establishment of regional anaesthesia. Alternatively, a combined spinal–epidural technique may be used, with a dense, but deliberately low, block being produced by intrathecal injection and anaesthesia gradually completed using the epidural catheter [16]. Unfortunately, in the case reported, epidural anaesthesia was considered to be contraindicated due to worsening thrombocytopaenia, with the potential risk of epidural haematoma formation [17]. Continuous spinal anaesthesia using a microspinal catheter was a technique that was not available to us [18].

Although no cause has been offered for the thrombocytopaenia present in this case prior to pregnancy, it was felt that the diminishing platelet count following admission was a manifestation of pre-eclampsia. Thrombocytopaenia has been reported as occurring in one-third of pre-eclamptic women [19]. It is commonly believed that, when the platelet count is above 100 × 109.l−1, clotting tests are unnecessary and all regional techniques can be performed while, below this value, clotting tests are often performed to help in the decision as to whether central regional blockade is safe [20]. Bleeding time has been suggested as an in vivo test of platelet function, but it is prone to inter- and intratechnician variability [21]. A potentially more useful test is thromboelastography, which provides a global assessment of haemostatic function, assessing the interaction between platelets, the coagulation cascade and fibrinogen [22]. Unfortunately, thromboelastography was not available to us. Epidural and spinal haematomata are rare following regional anaesthesia in obstetric practice, but when they do arise, it is often in the presence of a bleeding diathesis or anticoagulant drugs [23]. Although spinal haematoma has been reported following single-shot spinal anaesthesia [24], insertion of a narrow-gauge spinal needle is less traumatic than insertion of an epidural catheter and can be considered when the benefits outweigh the risks of a general anaesthetic [25]. When such a block is performed, any prolongation in the expected duration of the sensory and motor block should prompt urgent neurological referral and investigations to exclude the presence of a haematoma.

The sudden decrease in blood pressure seen in this case may have arisen as a result of the chronic vasoconstriction and relative hypovolaemia seen in both systemic sclerosis [10] and pre-eclampsia [26]. Again, an epidural or spinal catheter technique with gradual establishment of anaesthesia and sympathetic block would prevent sudden hypotension, although avoidance of aortocaval compression, cautious intravenous fluid infusion and prompt use of vasopressor agents will ameliorate any decrease in blood pressure seen with spinal anaesthesia [20].

General care of the patient with systemic sclerosis during anaesthesia should include careful positioning, preferably with the assistance of the conscious patient, padding to avoid pressure necrosis, wrapping of limbs for protection, warmth to avoid vasoconstriction and a warm operating theatre. Difficult venous access may require the insertion of a central venous cannula. A pulmonary artery catheter may be of value if cardiac involvement or pulmonary hypertension exist. Non-invasive blood pressure monitoring is to be preferred to arterial cannulation and prolonged application of a pulse oximeter probe to one digit should be avoided.

In conclusion, we have described the successful use of spinal anaesthesia for Caesarean section in a woman with systemic sclerosis and pre-eclampsia, while accepting that, when not contraindicated, incremental spinal or epidural anaesthesia using a catheter technique would have been preferable. Careful pre-operative assessment is required to determine the multisystem involvement of systemic sclerosis and allow formulation of an anaesthetic plan. Consequently, when possible, such patients should be seen at an early stage in pregnancy to discuss analgesia and anaesthetic management.