A deadly combination of AIDS, TB and cardiac tumour

Abstract

Immunocompromised status in AIDS makes differential diagnosis of any symptom very difficult for a clinician. Sharp clinical judgement and plenty of investigations may be needed to reach the diagnosis, as in this case. We hereby present a case of AIDS and active tuberculosis (TB) under treatment. The patient developed acute onset multifocal neurological symptoms following an episode of fever and diarrhoea. The MRI scan revealed numerous large cerebral infarcts. On investigations to evaluate brain infarcts, we made a diagnosis of left atrial cardiac tumour. Association of cardiac tumours with AIDS has only been rarely reported. It is uncertain if these can be opportunistic tumours in AIDS. The patient successfully came out of this deadly combination of diseases, viz AIDS, TB and large brain infarcts due to atrial tumour; with almost complete recovery.

Background

Immunocompromised status in AIDS makes the differential diagnosis of any symptom very difficult for a clinician. Lot of brain storming and plenty of investigations may be needed to reach the diagnosis.

Diagnosis of left atrial tumour in this case was an unexpected one. AIDS is known to predispose to opportunistic tumours but cardiac tumour is usually not encountered. This association has been previously reported very rarely in the medical literature.

Case presentation

A 41-year-old lady presented with transient memory loss, irrelevant talk, visual disturbances and left-sided weakness since 2 days. She had acute onset of loose motions, high fever and dehydration 4 days prior to admission; for which she was hospitalised in a small peripheral nursing home. There, she received intravenous fluids and other treatment. She recovered from diarrhoea and fever but developed the above-mentioned neurological symptoms. All these neurological symptoms were fluctuating in their occurrence and severity.

Her neurological examination revealed reduced power on left side (motor grade 3), some limping, occasional disorientation, confusion and blurred visual fields. Examination of cranial nerves was normal. Her vitals were normal.

She suffered from pulmonary tuberculosis 12 years prior to admission for which she took complete ATT (anti-TB treatment). She had recurrence of pulmonary tuberculosis 2 years prior to admission for which she took complete re-treatment and was cured again.

Three months prior to admission, she had another recurrence of pulmonary TB. Her x-ray chest showed a right lower lobe patch and she was found to be sputum AFB (acid-fast bacilli) positive. She was put on re-treatment regimen again after setting the mycobacterial culture and explaining the possibility of drug resistance. The culture later grew drug-sensitive Mycobacterium tuberculosis complex. She was responding well to ATT and was adherent to the treatment.

She was detected seropositive for HIV-1 8 years ago but had never done a CD4 count, had never considered ART (antiretroviral treatment) and was in a mental status of denial of the existence of HIV infection. She had to be frequently motivated for acceptance of the disease and further evaluation. She agreed to check CD4 count 13 months prior to admission (just before the completion of her first TB retreatment) and it was found to be only 12 cells/mm³. She was promptly started on Lamivudine, Zidovudine and Efavirenz. The risk of IRIS (immune reconstitution inflammatory syndrome) was explained. The patient was adherent to ART and ATT at the time of presentation and the CD4 count was 183 cells/mm³ after 13 months of regular ART.

Investigations

The MRI scan of brain showed a large right temporoparietal acute infarct showing restricted perfusion and appearing hyperintense on T2/flair image and hypointense on T1W1 with effacement of the cortical sulci. There were areas of iso to hyper intensity on apparent diffusion coefficient (ADC) suggestive of areas of early subacute infarcts/reperfusion (figure 1 shows the cerebral infarcts marked by arrows). Similar areas of restricted diffusion were noted in right anterior cerebral artery (ACA), middle cerebral artery (MCA) watershed region, caudate nucleus and areas of corona radiata suggestive of acute infarcts. Similar infarct was also noted in left occipital region (figure 2 shows the multifocal infarcts marked by arrows). The opercular branch of right middle cerebral artery showed loss of flow void suggestive of possible thrombus or embolus.

Figure 1.

Large cerebral infarcts.

Figure 2.

Multifocal cerebral infarcts.

Following investigations were therefore ordered to evaluate for source of emboli and also the possibility of vasculitis.

• TORCH Panel (antibodies against toxoplasma, rubella, cytomegalovirus, herpes simplex virus 1 and 2)

• Carotid Doppler

• Two-dimensional echocardiography

The TORCH panel was negative. Both common carotid arteries were traced with Doppler from their origin to bifurcation. They were normal in calibre and free from thrombi.

The echocardiography showed a large pedunculated homogeneous mass attached to the roof of left atrium measuring 24×12 mm and partially prolapsing through the mitral valve (figure 3 showing the mass in left atrial cavity marked by arrows). The mitral leaflets were normal, intra-atrial and intraventricular septae were intact. No intracardiac thrombus was found.

Figure 3.

Large pedunculated left atrial tumour.

Differential diagnosis

On initial presentation (prior to MRI), we considered the following differential diagnosis:

The neurological symptoms appeared after an episode of dehydration. This prompted us to consider three possibilities.

• Osmotic demyelination syndrome due to rapid correction of hyponatraemia was considered as a possible differential diagnosis. This normally evolves subacutely and primarily affects pons but may present acutely and affect other parts of the brain. We confirmed that the patient did not have any symptoms of hyponatraemia and her electrolytes were normal during her prior hospitalisation for diarrhoea.

• High fever, diarrhoea and acute neurological symptoms may indicate an infective aetiology affecting the brain and gastrointestinal tract together. Opportunistic infection such as mycobacterium avium complex (MAC) and possibility of sepsis with small cerebral abscesses were considered in view of persistently low CD4 count.

• Dehydration can precipitate an ischaemic episode in a predisposed individual and hence this was considered.

The neurological symptoms were waxing and waning and fluctuating in their occurrence and severity. This prompted us to consider two possibilities.

• Vasculitis: varicella zoster virus, HIV and cytomegalovirus, all can cause acute vasculitis with ischaemic symptoms/stroke. All were possible in this case and were required to be ruled out.

• Ongoing thromboembolic phenomena in the brain

Other possibilities could have been SOLs (space occupying lesions) such as tuberculoma and cerebral lymphoma. These are known to occur in the setting of AIDS and active TB. These would however typically present with convulsions. Acute paralytic stroke/memory loss would be possible with acute oedema/haemorrhage in these lesions.

Patient had no meningeal symptoms/signs and meningitis was therefore unlikely. TB meningitis would typically present with cranial nerve involvement. However, vague presentations are likely in an immunocompromised host.

Treatment

The patient was given low-molecular weight heparin and was put on warfarin later. She was considered high risk for any kind of surgical intervention in view of inadequate control of HIV, active TB and large sized cerebral infarcts. Also, the patient who was feeling depressed due to major illnesses one after another, did not consent even for discussing the possible risk and benefits of surgical excision of the tumour.

The ATT regimen was later completed and the TB got cured successfully with total 8 months of re-treatment regimen.

Monitoring international normalised ratio (INR) and adjusting dose of warfarin was a difficult task in view of concomitant ATT and ART due to very complicated drug interactions. Rifampicin reduces efavirenz levels by 25%. Efavirenz also being a potent inducer of Cytochrome P450 enzyme reduces the levels of warfarin. On the contrary, isoniazid increases the effect of warfarin. Rifampicin is known to interact with warfarin by reducing its levels. An option of changing to rifabutin-based ATT regimen was not considered as the TB treatment was ongoing; the patient was responding well and also there was documented drug sensitivity to rifampicin. Also, rifabutin is known to interact with warfarin in a similar way to rifampicin but to a lesser extent. Maintaining adequate anticoagulation was successfully accomplished with close monitoring of INR and suitable adjustments of warfarin doses.

Outcome and follow-up

The patient responded very well and relatively quickly over the next 3 months. She is now comfortable with almost full recovery of hemiparesis and total recovery from memory loss. Some visual disturbances persist in form of lost vision in the lateral field of left eye. She is afraid of the resulting partially tunnelled vision as it may predispose to accidents, especially on busy roads.

The incident did make her nervous and depressed initially. However, very strong moral support by the husband (who is HIV-negative) and the medical team encouraged her to come out of it successfully.

Her CD4 count was 178 at the end of her TB retreatment, which got completed 5 months after the tumour diagnosis. It had failed to rise sufficiently, in spite of 18 months of ART. A viral load was therefore measured and it was 145 copies/ml. This was considered as an unsatisfactory response in view of the current guidelines which define treatment failure as failure to achieve a viral load of <50 copies/ml 6 months after starting ART or following viral suppression to <50 copies/ml and a viral load rebound to >400 copies/ml on two consecutive occasions.¹ The resistance testing was not done due to non-availability of the test. After consultation with an expert on HIV and thorough discussion with the patient and her husband, she was switched over to a combination of Tenofovir, Emtricitabine and ritonavir boosted atazanavir.

INR has been currently maintained at 2.2. The viral load was repeated, 3 months after changing the ART regimen and it is less than 20 copies/ml. The CD4 count was evaluated recently and it has jumped to 340 cells/mm³ indicating favourable therapeutic response to the changed ART regimen.

She still refuses to give any consideration to resection of the atrial tumour.

Discussion

Atrial tumours are very rare and include myxoma, lipoma and rhabdomyoma. Though a confirmatory tissue diagnosis was not possible in this case due to patient's refusal for any surgical intervention, the echo characteristics were considered as sufficient evidence for diagnosis of myxoma. Other rare cardiac tumours such as lipoma and rhabdomyoma are not pedunculated and usually infiltrate the muscle of the heart.

Atrial myxoma is a benign tumour that commonly arises from the atrial septum or roof, usually on the left side and grows slowly.² Its mere presence can lead to blood flow obstruction that may result in dyspnoea.³ However, thromboembolic event is a more common clinical presentation.

The first report of atrial myxoma in a HIV-positive lady was published by Shaw and Mclean in 2000.⁴ This patient had a postembolic myocardial infarction. Normal coronaries were later proved by angiography. The myxoma was diagnosed while evaluating a syncopal attack 3 years later. The tumour was resected surgically inspite of low CD4 count. The postoperative period was uneventful and the patient survived well with ART. Another case of atrial myxoma in a patient of congenitally-acquired HIV has been reported.⁵

This case is probably only the fourth case of association of AIDS and atrial myxoma reported in the medical literature.

Although opportunistic benign or low-malignant tumours have been described in AIDS, most of these have been related to Epstein-Barr virus infection.^6–8 Any firm association of atrial tumours with AIDS has not been established.

Learning points.

• In the setting of AIDS, plenty of differential diagnoses need to be considered in case of a stroke which is a great clinical challenge.

• Association of AIDS and atrial myxoma has been very rarely reported before.