Tuberculosis in a prison, Jeddah, Saudi Arabia, July 1993-March 1995

After recognizing four extra pulmonary tuberculosis cases among prisoners hospitalized in a Jeddah hospital, an additional 40 prisoners with pulmonary TB were identified from TB surveillance between July 1993 to December 1994. An epidemiologic investigation was begun to determine the reasons for the large number of TB cases among prisoners in a prison in Jeddah.

The prison has two sections: a correctional institute and a general prison according to the crime. One clinic serves both sections of the prisons. prisoners with suspect TB are referred to the Jeddah TB Center for 'chest x ray, tuberculin skin testing, and sputum examination. Those with acid fast bacilli on sputum examinations or Mycobacterium tuberculosis on culture are referred to the Chest Hospitals in Taif for treatment. New prisoners had not been screened for TB.

To find cases we reviewed patients' records of the two hospitals used by the prison and the Jeddah TB Center. A pulmonary TB case was defined as a prisoner who developed a cough illness with a sputum smear positive for acid fast bacilli and extra pulmonary TB cases as an illness with TB granuloma demonstrated by histology from July 1993 to March 1995. Prisoners with TB were interviewed in the prison or Chest Hospital.

From July 1993 to February 1995, 53 cases of pulmonary (49 cases) and extra pulmonary (4 cases) TB were diagnosed among prisoners (incidence rate= 456 per 100,000 per year). TB cases had been detected in the correctional institute since the beginning (July 1993) of the study period and the first seven cases had onset of illness more than one month after imprisonment (prison-acquired). In contrast, no TB cases had been discovered in the general prison during the first five months of the case review and one of the first three cases had onset less than one month after imprisonment (community-acquired). For all TB cases time between imprisonment and onset of symptoms ranged from 0 to 1162 days (median 216) and 87% were prison-acquired.

Referral from the prison clinic for TB diagnosis was delayed from 31 to 65 days (mean 54) after onset of cough. The risk of TB increased with crowding in the ward (Chi square for trend = 5.1, p< 0.05) and time spent in prison (p< 0.01, ANOVA). When compared to control prisoners selected at random from all prisoners, prisoners with diabetes mellitus (Odds ratio [OR]=16, 95% Confidence Interval [CI]7-37) and smokers (OR=2.9, 95% CI 1.1-7.9) had a greater risk of TB.

We screened 297 prisoners using chest film sputum smears and tuberculin skin testing. Prisoners were selected because they were in a ward with known TB cases or because they had a risk factor for TB (HIV positive, diabetes mellitus, chronic renal failure, hematological disorder). Twenty-six previously undetected TB cases were identified (8754 per 100,000). The mean PPD reaction was 7 mm with a range from 020 mm among all screened prisoners. The rate of tuberculin positively

increased with increasing months of imprisonment (R= 0.27, 95% confidence limits 0.08-0.44). Screening of guards and social workers (18) detected one guard and one social worker from the Correctional Institute with pulmonary TB.

Prisons throughout the world represent a situation where risk factors for acquiring tuberculosis (TB) are common and require special attention for TB control and prevention.) [1,2) The prisoners themselves come from groups of people in the community that are at relatively high risk of TB. In the prisons relatively crowded conditions and close contact between prisoners increase the risk of TB transmission.[3]

Many of the factors previously noted in other prisons were demonstrated as contributing to TB in this prison. These include crowding and length of time spent in the prison, smoking, and underlying chronic disease.[4] Prisoners may have been selected from a high risk social groups in the community. However, this could not be shown because prisoners were not screened when first imprisoned. Transmission had probably been ongoing in the correctional institute before July 1993, whereas the community-acquired case among the initial cases in the general prison suggests that transmission there was more recent. The lack of screening of new prisoners and the delay of diagnosis of TB through the prison clinic were important contributors to this outbreak. In addition the screening indicated that there was a significant reservoir of undiagnosed pulmonary TB among the prisoners.

The problem of TB in prisons must be dealt with by increasing awareness of prison medical staff and authorities. To eliminate introduction of community-acquired TB, all new prisoners and prison employees should be screened with TB skin tests and chest x ray.[2] In prison , a more comprehensive screening of existing prisoners and staff is indicated. To prevent spread to the community, similar screening should also apply to released prisoners, especially those in contact with TB cases. Surveillance should include full access of prisoners to clinic services and heightened awareness of TB among clinic staff to promptly identify any new case of TB developing among prisoners. As soon as TB is diagnosed treatment should be started and contacts evaluated.