Long term valve-related morbidity following open-heart surgery at the Kenyatta National Hospital

To determine the incidence of valve morbidity associated with post-operative valve patients at the Kenyatta National Hospital, Nairobi.

A combined retrospective and prospective study. Restrospective period from 1973 to 31st July 1977 and prospective period from there on to 31st July 1998.

Cardiothoracic outpatient clinic at Kenyatta National Hospital, Nairobi.

All valve replacement patients attending the cardiothoracic outpatient clinic on follow up following discharge from the ward.

Morbidity was assessed by incidence of: stroke, thromboembolism, valve thrombosis, bleeding episodes, valve endocarditis and re-operation for valve failure.

For all valve surgery combined the overall follow up rate equalled 476.3 patient years. The linearised occurrence rate for thromboembolism for all valves was 1.04% per patient year. For the mechanical valves this incidence was 1.32% per patient year. The stroke free rate at one year and five years was 97.7% and 95.9% respectively. The linearised incidence rate for stroke was 0.84% per patient year for all valves. The incidence of bleeding episodes, excluding menorrhagia, for mechanical valves was 0.79% per patient year. Three episodes of valve thrombosis were observed in this study group, all occurring with mechanical valves. For mechanical valves the incidence equalled 0.79% per patient year. The incidence of prosthetic valve endocarditis equalled 0.42% per patient year for all valves combined, or 0.53% per patient year for mechanical valves. Three patients in the series had redo surgery due to valve failure (0.63%). The commutative occurrence of valve related morbidity for the study population for all valves analysed was 2.73% per patient year.

The incidence of the various valve morbidity at the Kenyatta National Hospital are below the internationally quoted figures in the literature for other centres. The high incidence of drop out to clinic follow up is likely to be the main reason for this with a lot of morbidity occurring outside the hospital environment and therefore not documented. The exact estimation of this incidence is difficult to assess. To help ascertain this occurrence an accurate patient database with accurate (real time) contacts needs to be put into place.