Outbreaks of kerato-conjunctivitis caused by adenoviruses or enteroviruses occur repeatedly as the main viral conjunctivitis in Japan. In particular, the nosocomial infection in university clinics with hospitalization facilities exert a strong impact upon outpatient care and hospitalization and surgical treatment. However, in the present context, the frequency of nosocomial infection, its background and the countermeasures have not been precisely identified. We have carried out a questionnaire survey on nosocomial infection of viral conjunctivitis in Japanese university hospitals during the last ten years.

1. Current situation of nosocomial infection
     This survey was carried out by mailing questionnaires to 93 ophthalmologic departments in university hospitals throughout Japan in order to study nosocomial infection that occurred during approximately ten years from 1993 to June 2002. As a result, we obtained replies from 62 hospitals (response rate: 66.7%). Over the past decade, 49 hospitals (79.0%) experienced nosocomial infections with viral conjunctivitis, of which the number of those which had repeatedly experienced was 13 (the infection occurred two times in nine hospitals and three times in four.). According to the replies to this survey, all the hospital-based infections were caused by adenoviruses. The number of hospitals which experienced outbreaks of the infectious from 1993 was 3~5 per year during five years from 1993 to 1997, and about 10 during the recent five years after 1997,infection broke out almost every month except in February. As far as the places where infection occurred are concerned, infection in hospital rooms only occurred in 33 hospitals (70.2%), simultaneous infection in hospital rooms and during ambulatory care occurred in 12 hospitals (25.5%), and infection during ambulatory care only occurred in 2 hospitals (4.3%). As regards the number of infected individuals, it was below 10 in 16 hospitals (32.7%), 10~30 in 16 hospitals and over 30 in 13 hospitals. Infection affecting health care workers was observed in 28 hospitals (58.3%). The persons who initially contracted conjunctivitis were hospitalized patients in 30 hospitals, outpatients in 12 hospitals, doctors in 3 hospitals, and health care workers in one hospital.
     As far as the route of infection is concerned (including definite and assumed routes), it was medical equipments in 28 hospitals and hospitalized patients in 23 hospitals, and the doctors in 17 hospitals were assumedly the route of infection. Medical equipments were as follows: aplanation chips in 14 hospitals, mirrors in 8 hospitals and stenopic microscopes in 3 hospitals.
     As regards the diagnosis, 12 hospitals carried out a clinical diagnosis only and 24 hospitals performed a virological diagnosis with Adenocheck,™ which is a rapid diagnostic tool. Virus isolation and culture were done in 7 hospitals, and adenovirus type 4, 8, 19, 37 were identified.
     The following measures were implemented to control nosocomial infection: compulsory discharge of infected patients in 32 hospitals, isolation of patients in private rooms in 29 hospitals, ban on new hospitalizations in 27 hospitals, discharge of all the members and closure of hospital wards in 16 hospitals. Five hospitals did not impose restrictions on hospitalization and discharge. The duration of the ward closure was less than 7 days in 6 hospitals and 8 to 16 days in 10 hospitals.

2. Future measures
     This survey revealed that nosocomial infection caused by viral conjunctivitis over the past ten years in Japanese university hospitals occurred in 79% of them. This clearly shows that hospital-based infection is quite frequent. During nosocomial infection, onset in physicians and health care personnel was observed in 58.3% of the surveyed hospitals. Four hospitals were assumed to be the initial infection site. Therefore as far as the prevention of nosocomial infection is concerned, it is important that the health care personnel including physicians, nurses, orthoptists and reception staff be fully aware that they may be the source of infection themselves. As regards routine crisis management of hospital-based infection, it is essential to focus on improving the consultation and treatment environment, to endeavor to detect promptly infected individuals and to take adequate action when dealing with these patients.
     Appropriate disinfection is required for aplanation chips and mirrors, which are medical equipments with supposedly the highest possibility of being the route of infection. At any rate aplanation chips and mirrors should be amply washed with water and dipped into a disinfectant, if adenoviral conjunctivitis is suspected. Aplanation chips float in the disinfectant and they should be occasionally stirred since the parts that are not impregnated are not disinfected. Besides the mirrors should be completely disinfected including the parts in contact with the ocular surface and those touched by the examiners. It is also necessary to disinfect well the stenopic microscopes after use with alcohol for disinfection. (They should be wiped two times.)
     On the other hand, only 24 hospitals performed their diagnosis with Adenocheck™. One may assume that this kit should be employed more frequently since early detection and early diagnosis are capital. The specificity of Adenocheck™ is 100%, but since its sensitivity is on the order of 60%, the involvement of Ad cannot be denied even when the results are negative. The laboratory staff should fully understand this fact, otherwise the response to Adenocheck™-negative individuals will be inadequate and the implementation of measures for the prevention of infection may be delayed. In addition virus DNA detection by means of the PCR technique and virus identification through isolation and incubation are also necessary to know precisely the actual number of infected individuals. Especially in case of nosocomial infection, it is important to identify the type of virus through some kinds of virological technique in order to know whether the infected individuals really share a common virus.
     Practically almost the hospitals surveyed took measures addressed to the hospitalized patients, such as compulsory discharge or isolation in private rooms of those who were infected, suspension of new hospitalizations and closure of hospital wards. Preventing nosocomial infection is the best, since this involves a sense of mistrust toward the medical profession and hospitals and the economic burden caused by the closure of hospital wards. Nevertheless the survey shows that in the event of an outbreak, the most important is to contain the spread to the minimum through early detection and early diagnosis and to show a sincere response to the patients.
     This questionnaire survey shows that nosocomial infection of adenoviral conjunctivitis occurs every year with a fairly high frequency in Japanese university hospitals. This is a major societal problem. Indeed one-third of the university hospitals closed wards. This survey has led us to consider that it is extremely important, on the one hand, to be aware that nosocomial infections may occur in future at any time and in any hospital and to work on infection prevention, and on the other hand, to implement early detection and early diagnosis and to rapidly take countermeasures in order to prevent the spread of the infection at the initial stage.