A Thirty year old primigravida immigrant presents with florid pulmonary oedema during Mid Cavity Forceps Delivery due to mitral stenosis.

S. McAree,1 C.Johnstone, 2 and G. Irvine 3
1. Specialist Registrar Anaesthetics, Victoria Infirmary, Glasgow, UK.2 Consultant Anaesthetist,Crosshouse Hospital, Kilmarnock, UK. 3. Consultant Obstetrician Ayrshire Maternity Hospital, Crosshouse, Kilmarnock. UK.
Correspondence to: Dr S. McAree Department of Anaesthetics, Victoria Infirmary, Langside Road, Glasgow G42 9TY. UK. simonmcaree@doctors.net.uk

Abstract
With an increase in the level of immigration to the United Kingdom, there are more challenging cases presenting to the clinician with language and cultural differences as well as old friends revisited in terms of pathology. In this case we present a 31 year old primigravida North African immigrant who managed to evade a diagnosis of Mitral Stenosis until presenting with acute pulmonary oedema during a Mid Cavity Forceps Delivery (MCFD) under epidural anaesthesia. This was a late diagnosis confounded by language barriers and a return of a classic condition perhaps more read about than actually seen in clinical practise. Issues related to the management of mitral stenosis are discussed along with some reflection on difficult communication issues, which exemplify the findings in the recent CEMACH 2007 report1.

Key Words: Mitral Stenosis, Communication, Immigration.

 

A slim 30 year old primigravida North African immigrant who could not speak any English booked in at 18 weeks pregnancy under the midwifery service with a totally unremarkable past medical history taken via a French interrupter including denial of rheumatic fever, tuberculosis and thrombo-embolism. The only medication being taken was prophylactic folic acid and there were no drug allergies. She had had been working as a cleaner prenatally. Her blood pressure (BP) at booking was 98/60 mmHg with no peripherial oedema present. It was noted that obtaining the history was difficult and perhaps an Arabic translator would be better in the future.

Her pregnancy proceeded without difficulty to thirty seven weeks whereupon she describes shortness of breath on exertion and a dry cough at her routine check via an Arabic interpreter. No further symptoms were offered. Symptoms relating to thromboembolic disease were specifically enquired about but denied. The lady was then reassured that some shortness of breath was to be expected at this point in pregnancy. She was normotensive but no examination of her chest or precordium was carried out or referral made to the medical staff.

She was admitted to the midwifery labour unit at term where examination revealed a well looking, normotensive woman. Palpation of the abdomen revealed an active fetus in a long lie with a cephalic presentation and normal urinalysis. It was then that the interpreter announced that the lady would prefer an epidural as her first choice analgesia during labour and so the patient was then transferred to the medical labour ward and an uncomplicated epidural insertion ensued.

An uneventful test dose of 2.5 mls 0.5% Bupivicaine was followed by 10mls 0.125% Bupivicaine and 100 micrograms of Fentanyl. This was complicated by hypotension (74/43 mmHg) associated with a foetal bradycardia treated with left lateral positioning, a fluid bolus of 1000ml Hartmans solution and ephredrine 6mg. The foetal heart responded well but the maternal response to this was a modest increase in BP (94/39mmHg) and a persisting tachycarda of 120. The continuous 0.1% bupivicaine and 2 micrograms.ml-1 fentanyl epidural infusion was discontinued until the blood pressure became more stable, some twenty minutes later. The whole episode was, however, asymptomatic. An hour later the patient developed a dry non-productive cough. A simple linctus was prescribed following a normal chest examination.

Labour was augmented at six hours with syntocinon due to the development of grade two meconium and a history of a possible prolonged spontaneous rupture of membranes. Intravenous benzylpenicillin was also commenced at this time. This was associated with breakthrough back pain and an epidural top up (5 mls 0.25% Bupivicaine) was given. Again this was associated with a drop in BP (86/48mmHg), a tachycardia of 120, coughing and vomiting. There was a decrease in variability on cardiotocograph but this recovered quickly on lateral positioning.

At nine hours the coughing had became worse. Chest examination now revealed bilateral basal crepitations, which were felt to be coarse in nature. The patient however reported to the contrary by feeling well and denying feeling short of breath. The pulse oximetry (SpO2) remained normal. The patient again refuted a past history of tuberculosis and thromboembolism but the history had to be obtained by yet another Arabic interpreter over the telephone as the original one was stricken with “flu”. The fluid balance for labour at this time was 2000mls positive. A fluid balance chart was therefore commenced and intravenous fluids were restricted. Routine blood tests were normal. The diagnosis was felt to be unclear but presumed to reflect lung rather cardiac pathology. A chest X-ray was planned post natally. A third Epidural Top up towards the end of the first stage was uneventful and by ten hours the cough had settled.

A decision was made after eighty minutes of pushing with the onset of late decelerations to manage the delivery with a trial of forceps in theatre. At this point the patient felt exhausted, smelt of ketones and complained of thirst. The maternal HR was 135 and the BP 130/75 mmHg. There was a systolic murmur along the left sternal border, which was attributed to a hyperdynamic flow, and a few scattered bilateral basal crepitations. The epidural was topped up in theatre using twenty mls of 0.5% Bupivicaine over the course of thirty minutes to establish a block to T4 bilaterally to cold. Her BP was treated expectantly with Hartmans 1000mls running in over the 30 minutes as the patient appeared dehydrated. A further 500mls of gelofusin and two 50 mcg phenylephrine boluses were also used to maintain BP. The heart rate settled to 110 and there was little evidence of impaired gas exchange with SpO2 98% on air. A healthy baby was delivered by MCFD, with concurrent active pushing, at midnight. Umbilical cord bloods were satisfactory with an operative blood loss was 300mls.

Following the delivery of the placenta the patient’s oxygen saturation began to fall with a SpO2 93% on air and a respiratory rate increase from 18 to 26. The patient was not symptomatic and was initially reluctant to commence oxygen therapy. The heart rate increased again to 135 but the blood pressure was maintained at 120/70 mmHg. A high epidural block was excluded clinically with a sensory level at T4 bilaterally to cold. There was a continued increase in oxygen requirements over the next 15 minutes and a worsening tachypnoea, which had by now reached 34.

The patient was then transferred to the neighbouring High Dependency room in the labour suite and sat up. Oxygenation had to be maintained using high flow oxygen via a Trauma mask; the SpO2 having fallen to 73% on a Hudson mask. Chest examination revealed widespread crepitations throughout all zones with the addition of expiratory polyphonic wheeze. A prospective diagnosis of acute pulmonary oedema was made and Furosemide 10 mg was given (BP 96/50mmHg) slowly intravenously. A Chest X-ray (CXR) (figure 1) and ECG (figure 2) were requested. A urinary catheter and arterial line were also inserted. The CXR demonstrated a globular heart with widespread fluffy alveolar shadows consistent with pulmonary oedema and the ECG showed a sinus tachycardia with “p” mitrale. A Full blood count, electrolytes, clotting and Troponin were sent but the results were unremarkable.

 

A second bolus of Furosemide 10 mg was given 20 minutes later as her BP had improved after the first bolus and empirical antibiotics were also given at this stage. Over the following hour there was a good diuresis of 850 mls, and the respiratory rate and wheeae had settled. With persisting crackles in the lower third of the chest, a slowing diuresis and a residual epidural weakness, a further bolus of 20 mg Furosemide i.v. was given with continued improvement throughout the night.

An echocardiogram performed the following morning revealed severe mixed rheumatic mitral valve disease. Endocarditis prophylaxsis was completed with the addition of gentamicin. Oral Furosemide 40mg o.d was also initated and a referral to a Cardiologist made. The remainder of the post partum period was uneventful and the patient discharged without medication to be followed up in the cardiology outpatient clinic.

Discussion
In retrospect the diagnosis of mixed rheumatic valve disease with the first presentation in labour was a classic textbook case. An earlier diagnosis, however, would have enabled better planning of labour, an assessment of risk and appropriate management instituted. Our patient was asymptomatic and these patients usually tolerate pregnancy and delivery well and do not require invasive monitoring2. That aside perhaps pulmonary oedema could have been avoided or its impact reduced if the diagnosis was known sooner by avoiding the persistent tachycardia, careful attention to maintaining preload and treating the drops in systemic vascular resistance with a vasoconstrictor rather than ephredrine. In the second stage avoidance of prolonged pushing and considering an assisted delivery earlier. The most significant pulmonary oedema occurred when the uterus was contracting down and vena cava occlusion was released. This responded well to simple diuretic therapy, but judicious use of fluid pre-delivery and greater use of vasoconstrictors would have enabled better patient adaptation to the auto transfusion that accompanies delivery.

The incidence of Rheumatic Fever in the United Kingdom is extremely low and indeed antibiotic treatment of pharyngitis is now discouraged in terms of its prevention3. As a consequence mitral stenosis has become somewhat of clinical rarity. This is not true for other parts of the world where both conditions are much more prevalent. With immigration to the United Kingdom reaching over half a million in 20054, the number of immigrant cases presenting to the National Health Service will be rising as well as a much more varied disease case mix.

This case illustrates some of the difficulties in correctly diagnosing and treating trans-cultural patients. Firstly, the different pathologies encountered will result in broader differential diagnoses and secondly problems with communication, not just the language barriers but social, cultural and educational differences which exist too. Gaps in the communication develop, which are filled with unconsciousness assumptions. Time is a precious commodity and fully exploring details may not be practical in a busy clinic or indeed in labour where obvious clinical needs exclude long conversations.

In order to communicate with our patient we used a variety of methods: ad hoc with staff using French via her husband who translated the French into Arabic, one midwife who had a few words of Arabic to communicate directly and several commercial translators either over the telephone or in attendance at the antenatal clinics. Our efforts to provide consistent translation throughout the delivery were thwarted by an acute illness of the translator and it was very difficult for a labouring woman to speak over the telephone, so once again the translation had to go through a third party. Bilingual physicians and professional interpreters result in the optimum communication5,6,7 unfortunately this is not always available. Out of hours any preparation for interviews in a controlled fashion is impractical in the acute unpredictable nature of a maternity unit. Where instructions have to be given frequently the use friends and relatives to communicate although not recommended is inevitable.

The use of translators is not straightforward with many pitfalls8. For example, how do we know that our questions posed via the translator are accurate? What is Arabic for Rheumatic Fever for instance and would a translator without medical training have known this well enough to ask such a specific question? Would they automatically tell you that they did not know?

There is some guidance on how we as clinicians should conduct such interviews with translators to obtain the best history from the patient and therefore maximise clinical effectiveness of the consultation5,6,7. Unfortunately history taking is a finely tuned clinical skill, honed over many years of training. Using translators is an added complexity in this process with numerous “key points” to remember. Skill acquirement takes practice. As it is unlikely to become second nature after reading an article in a journal, has the time now come for training in bicultural interview techniques?

A further worrying feature which came to light during this case is that despite having respiratory symptoms during pregnancy, there was no note of a clinical examination of praecordium or chest at any point during her pregnancy. This is also highlighted in the recent CEMACH report that all migrant women should have a full medical assessment and cardiovascular examination at booking. This may be the patients own General Practitioner.

Medical errors are too common9, occuring in 11% of acute care inpatients10 and with communication cited as a major cause of medical errors11 it would put patients with transcultural barriers at particular risk. This is before we add in the broader medical problems of the immigrant population. In retrospect the delayed diagnosis in this case would be highly predicted and in keeping with the findings of the recent CEMACH report.

Conclusion

These sorts of challenges are set to continue. CEMACH has highlighted the excess mortality in immigrant groups very clearly and made recommendations. This case report has highlighted the difficulties that these groups can present. What is needed now is for the recommendations and learning points from this case to be put into practice at a local level. Are we ready to deal with these challenges?

1. G. Lewis (ed) 2007. The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving mother’s lives: reviewing maternal deaths to make motherhood safer – 2003-2005. The Seventh Report on Confidential Enquiries into Maternity Deaths in the United Kingdom. London: CEMACH.
2. Sarah Hughes The Obstetric Patient with Cardiac Disease in: Rachel.E Collis, Felicity Plaat and John Urquhart eds. Textbook of Obstetric Anaesthesia. Greenwich Medical Media Ltd. Published 2002:183.
3. Anonymous. Section 5: Antibiotics in sore throat Scottish Intercollegiate Guidelines Network Website address: http://www.sign.ac.uk/guidelines/fulltext/34/section5.html (accessed 08/04/2008).
4. The Office of National Statistics http://www.statistics.gov.uk/CCI/nugget.asp?ID=260&Pos=4&ColRank=2&Rank=320 (accessed 08/04/2008).
5. J. Poss and T. Beeman Effective Use of Interpreters in Health Care: Guidelines for Nurse Mangers and Clinicians. Seminars for Nurse Managers, 1999; 4:166-171
6. Flores G. The impact of medical interpreter services on the quality of health care: a systematic review. Medical Care Research and Review. 2005; 62: 255-297.
7. Michael Phelan, Sue Parkman, How to Do It: Work with an interpreter British Medical Journal 1995; 311: 555-557
8. Ebden P, Bhatt A, Carey DJ, Harrison B. The bilingual consultation. The Lancet 1988;1:347
9. K G M M Alberti Medical errors: a common problem British Medical Journal, 2001; 322: 501–502
10. Charles Vincent, Graham Neale, and Maria Woloshynowych Adverse events in British hospitals: preliminary retrospective record review British Medical Journal 2001; 322: 517-519.
11. Steven H Woolf, Anton J Kazel, Susan M Dovey et al. A String of mistakes: The Importance of Cascade Analysis in Describing, Counting, and Preventing Medical Error Annals of Family Medicine 2004; 2:317-26