Abstract
An awareness of diseases causing altered mental status (AMS) in particular localities could greatly facilitate patient management. This cross-sectional study evaluated 100 consecutive patients with AMS admitted to a hospital in Uganda. Patients were assessed by the Confusion Assessment Method. History, examination, available laboratory tests and patients' response to treatment were used to identify aetiologies. Our study included 58 males and 42 females: 82% were 16–50 years old and 38% were HIV-infected. The most common cause of AMS was infection (51.3%), with cerebral malaria and meningitis predominating. The aetiology was unidentified in 12%. The in-hospital mortality rate was 44%, with HIV infection being positively associated.
As infections and metabolic derangements, the most common causes of AMS in our setting, are mostly treatable with a relatively favourable outcome, critical evaluation, early intervention and improved investigative capacity would greatly improve patient outcome.
Introduction
The definition of altered mental state (AMS) extends from confusion to coma. Delirium, a clinical entity due to an underlying medical condition, is the most precise – an acute decline in attention and cognition which can be resolved with the treatment of the underlying disorder. 1,2
General hospital delirium prevalence is reported as 10–40%, 1,3 with mortality rates from 25–60%. 3,4 Infections and metabolic disorders were the most common causes in Ethiopia. 4 In the USA, the patients' histories and physical examinations were more diagnostically useful than ancillary investigations. 5
Patients of different ages, geographical and socio-economic settings present with different causes of AMS. With appropriate diagnosis and management, they can often be treated successfully, while delay in diagnosis and treatment may prove fatal. 2–4 Prompt evaluation and treatment are essential in order to decrease the associated morbidity and mortality. 1,2 A greater awareness of the diseases which can cause AMS in a particular locality would also greatly facilitate patient management. 4
Therefore, this study was undertaken in south western Uganda in order to determine the prevalence and causes of AMS in patients admitted to a general medical ward.
Methods
This observational study was conducted on the adult general medical ward of the Mbarara regional referral hospital, without interfering with the routine patient management. One hundred consecutive patients with AMS on admission from April to June 2007 were selected using the Confusion Assessment Method (CAM), which has a sensitivity of 94–100% and specificity of 90–95% in recognising delirium. 3,6,7 Patients who developed AMS after admission, or had a history of preceding trauma, were excluded.
Study patients were evaluated daily. The patient's history, a physical examination, available investigations (i.e., routine haematology and biochemistry, malaria screening, HIV screening, blood cultures, cerebrospinal fluid examination, chest X-rays and oximetry, as appropriate and depending on affordability), and the patient's response to treatment were used in order to make the diagnoses.
Demographic data were recorded. The relationships between variables were assessed using χ2 and P value tests. A P value < 0.05 was used as a basis of statistical significance. Risk ratios and confidence intervals were computed.
Results
Over the two-month period, 621 patients were admitted. Of these, 100 (16.1%) had AMS, 58 of whom were male, 38 HIV-positive and the majority were younger than 50 years.
Table 1 shows the frequency of the different identified causes of AMS. More than half (58%) had infections, predominated by malaria, sepsis and meningitis. The second most frequent cause was metabolic disorder (16%), with hypoglycaemia predominating, and in the majority of cases these were associated with the use of hypoglycaemic drugs. Of five patients with organ dysfunction, four had liver failure and one was uraemic. Of 14 patients with AMS secondary to cerebral hypoxia, seven had severe anaemia, four had hypoxaemia due to severe pneumonias and three had cerebral hypo-perfusion due to cardiogenic shock. Drugs contributed to 13 cases (13%), 12 of whom were intentional suicide attempts, mostly with organophosphates. Five patients had brain structural lesions – three had a stroke and one had a glioma. In some patients there was more than one cause of AMS.
Causes of altered mental status (AMS)
Table 2 shows the frequencies of AMS causes depending on HIV states. Of the 21 cases of malaria, 18 were HIV-negative. Almost all the cases of meningitis and encephalitis occurred in HIV-positive patients. All the cases of acute meningitis were due to Streptococcus pneumoniae, determined by cerebrospinal fluid tests. Chronic meningitis included TB and cryptococcal meningitis. Cases of encephalitis included probable herpes simplex, herpes zoster and cytomegalovirus, although laboratory confirmation was not possible. The one HIV-negative patient diagnosed with probable tuberculous meningitis was severely malnourished.
Causes of altered mental status (AMS) in the different HIV serostatus groups
The in-hospital mortality rate was 44%, with HIV infection being significantly associated (P value 0.01).
Discussion
This study demonstrates an on-admission AMS prevalence in the general medical ward of 16.1%, similar to previous studies. 1,3 Infections and metabolic derangements were the most common causes which is similar to those reported in the Ethiopian setting study. 4 The ≥65 years group had more sepsis compared to the younger groups (63.6% versus 5.6%).
Of the 21 cases of malaria, 18 were HIV-negative. HIV-1 infection is associated with a higher frequency of clinical malaria and parasitaemia, the association becoming more pronounced with advancing immunosuppression. 8–10 However, this is not reflected in our study, probably because of the small sample size. However, it could be due to HIV integrated care programmes, where all the patients are given cotrimoxazole prophylaxis and mosquito nets.
Neither HIV infection nor advanced age had a significant association with development of AMS but, again, this may be due to the small sample size of our study. None of the study patients had cancer co-morbidity or a previous history of delirium, so their significance in the development of AMS was not assessed.
The in-patient mortality rate of 44% was somewhat lower than that seen in the Ethiopian study. 4 Late presentation to hospital, limited resources and expeditious patient management, together with the severe HIV-immunosuppression of many patients, probably contributed to this high mortality.
This study has shown that the most common causes of impaired consciousness in the hospital admissions in our setting were infections and metabolic derangements. Both are potentially correctable, with a good outcome, if recognised and treated early. An emphasis on the education of health workers, an increasing awareness of the need for early diagnosis and treatment and an improved investigative capacity should improve the morbidity and mortality associated with this condition. Consistent postmortem examinations would be very useful, as only 12.5% of the diagnoses were definite and the cause of AMS was not established in 12%.