With the alterations in body physiology during pregnancy - serum calcium changes, orthostatic episodes and a changing centre of gravity - it would be expected that pregnant women are easily prone to bony injuries of all sorts, not least, spinal injury. But surveys have shown that the reverse could actually be the case (2). The voluntary reduction of physical activity, especially in itinerary, coupled with the self-imposed confinement to the more protective home environment, have been opined by Albright, et al, as possible explanations for this risk reduction in pregnant women (2). Worldwide, prevalence of spinal injury is 2 : 100 000 population per year; and more than 50% of them are aged 16 - 30 years, with a male : female ratio of 4 : 1 (17, 6). Road traffic accidents 35 - 50%, constitute the major aetiologic factor, followed by violent assaults 25 - 30%, falls 19% and recreational injuries 7% (6, 8, 1). The incidence in pregnant women separate from the overall female incidence is unknown, and there are very few published reports on this population of patients.
Our Centre, University College Hospital, Ibadan, is a major referral facility for neurological diseases for, possibly, a quarter to a third of Nigeria’s population. We admit 7 - 8 new, confirmed spinal injury cases monthly, an average of about 90 patients per year. For the eight years before 2002, only 3 cases of pregnant women with spinal cord injuries were recorded, mean 0.38 per year. However, in the two years thereafter, September 2002 to May 2004, 3 consecutive cases presented to our service - mean 1.5 per year - a 4-fold increase in incidence, representing about 1.67% of our spinal injury admissions within the period. All the 3 patients presented with cervical spinal cord injuries, two with incomplete quadriplegia and one with mild quadriparesis; and were all conservatively managed with the Gardner-Wells’ Tongs (GWT) traction. They all went into pre-term labour, had spontaneous vaginal delivery of normal live babies and made complete neurological recovery, respectively.
Case I. 25-year old, right handed, primigravid, Polytechnic undergraduate; passenger motor vehicular road traffic accident (RTA), immediate loss of consciousness lasting 2hours, subsequent inability to move any of the 4 limbs, associated generalized body pains and bi-sphicteric dysfunction. She was transported from a private clinic to our Centre about 9hours after the incident, without any spinal injury precautions, no cervical orthosis. She was pregnant - gestational age (G.A) 29weeks.
Clinical findings: Stable cardio-pulmonary status, mild pallor, Glasgow Coma Scale (GCS) score 14 (confused), normal cranial nerves, equal reactive pupils, moderate midline nuchal tenderness, global hypotonia, muscle power - upper limb groups 0, lower limb groups 2; left extensor and right flexor plantar responses, absent abdominal reflex; depressed muscle stretch reflexes, sensory level at the 4th cervical dermatome (C4) with diminished sensations to 8th thoracic dermatome (T8); lax anal sphincter, absent anal reflex, uterine fundal height 30/52. Spinal shock resolved after 72hours with diminished sensations elicited down to the 5th sacral dermatome (S5).
Investigations. Packed cell volume (PCV) 22%; plain spinal x-rays - loss of cervical lordosis, C4/C5 anterior 4mm subluxation, fracture of C4 spinous process; abdomino-pelvic ultrasonography - normal singleton foetus.
Diagnoses: 1. C4 traumatic incomplete quadriplegia (Frankel C) - Atypical central cord syndrome 2. C4/C5 anterior subluxation with C4 spine fracture 3. Anaemia in pregnancy 4. Mild (closed) head injury.
Case II. 30-year old, left handed, multigravid, petty trader, pedestrian motor vehicular RTA, associated altered sensorium but no outright loss of consciousness, severe headaches, bleeding frontal scalp laceration and weakness of all 4 limbs. She was transferred to our Centre from a private clinic 2 1/2hours after the incident with no spinal injury precautions. Her GA was 32weeks.
Clinical findings: Stable cardio-pulmonary parameters, mild pallor, easy irritability, GCS score 12/15, subtle left facioparesis, equal sluggish pupils, mild midline nuchal tenderness, global hypotonia, motor power 4- globally, left Babinski and right flat plantar responses, abdominal reflex globally present, exaggerated muscle stretch reflexes, intact sensations to S5, good anal sphincter tone, brisk anal reflex, bi-frontal scalp laceration 12cm long without underlying fracture, uterine fundal height 32weeks.
Investigatios: PCV 24%; plain x-rays - C4/C5 anterior 3mm subluxation unstable at flexion-extension (F-E) stress studies; abdomino-pelvic ultrasonography - normal singleton foetus.
Diagnoses. 1. C4 traumatic quadriparesis (Frankel D) 2. C4/C5 unstable anterior subluxation 3. . Anaemia in pregnancy 4. Moderate (closed) head injury - cerebral concussion and scalp laceration
Case III. 35-year old, right handed, multigravid, petty trader, pedestrian motor vehicular RTA, immediate loss of consciousness lasting 24hours, inability to move any of the 4 limbs, subsequently. Transferred to our Centre 16days after the incident from a private clinic without a diagnosis of spinal injury and of course, no spinal injury precautions.
Clinical findings (16 days post-trauma): Stable cardio-respiratory status, mild pallor, GCS score 15, normal cranial nerves, equal reactive pupils, no midline nuchal tenderness, global hypotonia, diminished motor power - upper limb groups 2 - 3, and lower limb groups 0 - 2+; weak flexor plantar response bilaterally, exaggerated muscle stretch reflexes, C4 sensory level with diminished sensation to S5, diminished anal sphincter tone, weak anal reflex, uterine fundal height 28weeks.
Investigations: PCV 22%; plain x-ray studies - straightened cervical lordosis, increased C4/C5 pre-vertebral soft tissue shadow, no spinal column translocation but F-E (stress) studies showed C4/C5 anterior 2mm subluxation on flexion correctable on extension; abdomino-pelvic ultrasonography - normal singleton foetus.
Diagnoses: 1. C4 traumatic incomplete quadriplegia without spinal instability. 2. C4/C5 prevertebral soft tissue contusion. 3. Anaemia in pregnancy.
All 3 patients were co-managed with the Obstetrics/Gynaecology and Haematology units. They were admitted into the spinal ward and, respectively had skull traction with GWT. Subluxation was reduced by GWT traction and maintained with 5, 4.5 and 4kg weights for a total of 42, 56 and 42days for Cases I, II and III, respectively. Occipito-parietal, sacral and calcaneal foam troughs were lodged under bony pressure points. Full laboratory work-up was done and the patients placed on analgesics, anti-coagulants (low molecular weight Heparin initially, and Dicumarol, later), ante-natal drugs with double dose haematinics, clean intermittent urinary bladder catheterization, chest and limb physiotherapy and other standard spinal injury protocols.
Motor and sensory functions progressively improved in all 3.
Case I. Went into spontaneous labour at 35weeks GA, with GWT in-situ, one hour after F-E studies on the last day of traction. Obstetric evaluation revealed the foetal head was in-perineum so, she was immediately transferred to the labour room where a spontaneous vaginal delivery of a 2.3kg live normal female baby (APGAR1 9) was accomplished, aided by an episiotomy. Post-traction F-E films confirmed satisfactory alignment and callus immobilization at the previous injury site. The tongs were removed 12 hours post-partum. Gradual ambulation was commenced and she was discharged home at motor power 3+ - 4 and 4+ - 5 for the upper and lower limb muscle groups, respectively, ambulating with minimal support. Her review 20months post-injury showed that she had made a full sensori-motor recovery, was back to all pre-morbid activities and in the 3rd trimester of another pregnancy. In the 24th month post-injury, she had another spontaneous vaginal delivery of a full term live normal male baby, 3.2kg weight, after a 5-hour labour, associated with severe lower abdominal and waist pains this time around.
Case II. Premature rupture of membrane in the 4th week of traction, at 36weeks GA, followed 30minutes later by a precipitate labour and vaginal delivery of a live normal female 2.6kg baby, APGAR1 8. Delivery was completed in an improvised labour room in the spinal ward with traction in-situ, and the patient with her baby transferred to the post-natal ward, subsequently. Adequate alignment and callus immobilization at the subluxation site was confirmed after 8 weeks on traction, and the tongs were removed 24 hours later. Motor power was 5 globally, and ambulation was promptly instituted. She was fully ambulant at discharge, and by the last review 14weeks post-discharge she was back to all pre-morbid activities.
Case III. Visible uterine contractions were detected 24hours after GWT traction insertion, at 36weeks GA with associated lower abdominal pains, and obstetric evaluation confirmed the first stage of labour. Delay in this stage necessitated augmentation with intravenous Pitocin infusion till vaginal delivery 9hours later, of a 2.45kg live normal male baby, APGAR1 7. The baby was transferred to the special care baby unit. The patient’s traction was continued until the 42nd day when F-E studies confirmed adequate callus immobilization, and the tongs were removed. On progressive ambulation with minimal support, she was discharged home at motor power of 4 and 4+ for the right and left limb muscle groups, respectively. Review in the clinic 6weeks post-discharge showed full sensori-motor recovery, ambulating without support and back to all pre-morbid activities.
Spinal injury management starts at the point of injury, through transportation to a health facility, till domiciliary discharge. At each of these points, neurological functions could be lost or previous gains reversed if standard protocols are breached or compromised. Secondary injuries could arise from re-injury at an immobilized unstable translocation site, cord oedema, hypovolaemia, hypoxia, anaemia, electrolyte imbalances, glycaemic alterations / other metabolic derangements or infections. Other documented causes of secodary injury include a fall in spinal blood flow, haematoma collection, release of neurotransmitters and toxic by-products, calcium influx into traumatized axons leading to the activation of proteases and phospholipases and generation of free radicals (16, 18, 4, 13, 7, 12). Without immobilization or splinting, an incomplete spinal injury could be converted to a complete injury, or one without neurological deficits could deteriorate to severe deficits, thereby changing the ultimate outcome, and the life of a patient. This becomes relevant when it is realized that none of our 3 pregnant patients received any form of spinal injury care before arrival in our Centre, not even for Case III referred from a private clinic 16 days after injury, in whom of course, a diagnosis was not made ab initio. This simply reflects the state of our country’s health system. There are no facilities on the highways for road traffic victims, neither are there national protocols for emergency rescue operations from natural disasters. It also appears that health care providers outside the Neurosurgical specialty seem not to be aware of the preliminary measures to safeguard spinal injury patients. This ignorance could play a huge role in the morbidity and mortality from these injuries nationwide, and in the actual number of patients that reach our Centre alive.
In investigating spinal injuries, plain x-ray studies, Computerized Tomography (CT) and Magnetic Resonance Imaging (MRI) are invaluable. Bony lesions are best outlined by CT, and to a great extent by plain x-ray studies; but with soft tissue, including spinal cord lesions, MRI remains the gold standard (19). We opted for the plain x-ray study in each of the cases because :
a) it was affordable and readily available
b) it was adequate for our evaluation
c) at the 3rd trimester of gestation, this load of irradiation is still considered safe
d) CT at a much higher cost and load of irradiation would not give any extra information going by the preliminary x-ray evaluation; and
e) MRI was neither readily available nor affordable for an average Nigerian, and in retrospect, would not have significantly altered our proposed line of management.
It should however be emphasized that Flanders, et al, have affirmed that there might not be any correlation between the radiologically observed bony injury and the degree of spinal cord injury/neurological deficits; such that in ideal circumstances, as soon as deficits are detected, a detailed evaluation with MRI should be done to make an accurate diagnosis of the cord lesion, documented to have a dependable predictive value for outcome (9, 2, 15, 14). This would thus aid the decision on the appropriate treatment modality to give the patient the optimum chance for neurological and functional recovery.
Depending on the nature and associations of the injury, and the experience and preference of the surgeon, treatment of cervical spinal injuries could be operative or non-operative (conservative), or a combination of both. In our cases we used closed skull traction with GWT because we could achieve adequate reduction and re-alignment, it was readily available, and we have a long experience with this modality. The Halo brace jacket could have been helpful especially for Case III who had no spinal column translocation, but its non-availability in our environment and high cost of importation precluded its consideration, coupled with the possible difficulties with its application in the face of progressive abdominal enlargement of 3rd trimester pregnancy. Otherwise, early mobilization of the patient out of bed is the Halo brace’s cardinal advantage over all forms of spinal traction, which in relevant cases would equally make the operative modality preferable to a conservative approach.
Bed confinement on GWT traction, which lasts for several weeks, carries with it many risks and complications more especially in the pregnant patient viz. deep venous thrombosis (DVT) with thrombo-embolism, pressure ulceration, orthostatic hypotension, hypostatic pneumonia, urinary tract infection (UTI) and calculi formation (3). Rigid adherence to standard protocols helped us avert most of these problems, though two of our patients developed urinary tract infections and were promptly investigated and appropriately treated. All 3 patients, suffered two bouts of malaria each, before discharge and were treated with appropriate anti-microbials but none had DVT, embolic phenomenon or pressure ulcers. A breach in the protocol of clean intermittent bladder catheterization was thought to be responsible for the UTI.
An early operation for open reduction and internal fixation prevents the prolonged immobilization attendant in traction because it ensures early ambulation, but studies have not universally demonstrated its superiority over traction in the long term outcome from spinal cord injuries (5, 10). Albright, however, advocated a strong consideration for caesarian section delivery in order to prevent a re-injury of the cord during the process of labour, if the integrity of the spinal immobilization could not be guaranteed (2). An interesting finding in these cases was the consistent presentation of our patients at the 3rd trimester of gestation. Could it be a negation of Albright’s hypothesis alluding that there is a lessening of itinerary and voluntary withdrawal to the safer confines of the home by pregnant women, which he posited was part of the reason for their low incidence of spinal injury (2)? Could it be that Albright is right, but that as the anxiety of the earlier months of pregnancy wears away pregnant women become more daring and itinerant? Or is it just a curious co-incidence that all 3 patients happened to be in their 3rd trimester of gestation at the time of trauma?
Less bewildering, however, is the fact that all the incidents resulted from road traffic accidents, the major aetiologic factor in spinal injuries worldwide 35 - 50% (6, 8, 1). Also, the cervical spine confirmed by various studies as the most frequently injured region (42 - 50%), was the involved region in all our cases. Studies have put injury at other sites as 31% and 27% for the thoracic and lumbosacral spines, respectively (11). The obstetric aspect is noteworthy too. All 3 went into pre-term labour between the 35th and 36th week GA, none of them attaining the full term of 40weeks. Stress from trauma is known to be a major factor in premature uterine contraction and has been implicated in spontaneous abortion and premature parturition. This possibility coupled with the stress and inconvenience of skull traction could be responsible for the pre-term delivery unless during trauma, the CNS elaborates any specific cytokines yet to be identified, that have oxytocic effects. Very likely, too, the premature uterine contractions could be part of the autonomic dysreflexia that usually accompanies spinal cord injuries.
Despite their neurological deficits, all our patients went into spontaneous uterine contractions, proceeding to vertex delivery, save for Case III who required Pitocin augmentation for poor progress of labour. Her case might not be attributed to the spinal injury since it is well established that even neurologically intact women could have weak uterine contractions and delay at any stage of labour attributable to non-neurologic causes. Normal labour was, however, possible in the other patients because the physiology of uterine contraction revolves around the action of the posterior Pituitary gland hormone, Oxytocin, synthesized in the supra-optic nucleus of the hypothalamus, and not directly on the uterine nervous supply. The function of the diaphragm and anterior abdominal wall muscles, nonetheless, are essential in the inevitable valsalva manoeuvre required to effectively bear-down at parturition. And then, more revealing is the fact that all 3 patients went into labour apparently unaware of the onset of labour pains. Head-in perineum and sudden rupture of membrane could all have been preceded by the gradual onset of labour pains, progressing over some hours to the gross uterine dilatations obvious on the anterior abdominal wall. Sensory deficits, at C4 level, could explain this remarkable regional anaesthesia. Pain fibres passing in the dorsal nerve roots and ascending in the contralateral spinothalamic tracts, with the rest in the ascending dorsal columns were all seemingly affected in the cord injury that also produced motor deficits, leading to quadriplegia. Spinal cord injury, therefore, appears to provide a non-pharmacologic regional analgesia that abolishes the sever pains of labour without impairing the normal process of parturition.
They all completed labour with GWT traction in-situ, delivering live normal babies with good APGAR1 scores (7 - 9), scores known to reflect the intrauterine environment of the foetus during the transition period of labour. The least score of 7 in Case III was likely a reflection of the mild foetal distress induced by 9hours of augmentation with Pitocin, since all the babies had maximum scores of 10 at APGAR5. But most importantly, all the patients had good outcome scores - complete sensori-motor recovery, full ambulation and return to pre-morbid activities, which affirms that optimal results and full neurological recovery could be attained with conservative management of spinal injuries with GWT traction, even in pregnant women. The first patient actually went ahead, within 24months post-injury, to have another pregnancy, with a spontaneous vaginal delivery of a normal baby, at full term.
It appears that the presence of GWT traction at does not lead to any additional maternal or foetal distress unexplained by the normal events of labour. If anything, the injury itself brings about a regional analgesia that confers a pain-free labour and does not jeopardize parturition.
In situations when surgical treatment is not readily feasible or financial difficulties prevail, adequate spinal reduction and immobilization with optimal neurological recovery could be achieved in pregnant spinal injury patients using the GWT traction, provided that standard spinal injury management protocols are observed.