Summary report on Hong Kong influenza in Japan

Bull. Org. mond. Sante 11969, 41, 353-359Bull. Wld Hlth Org. Summary Report on Hong Kong Influenza in Japan HIDEO FUKUMI 1 In the 2 months following the onset of the influenza epidemic in Hong Kong in July 1968, there must have been many importations of the virus into Japan both from Hong Kong andfrom other countries subsequently involved in the epidemic. Over a dozen such incidents were reported in Japan and confirmed serologically or by direct virus isolation. Although they must have seeded the virus at least in and near the ports ofimportation, a true epidemic of Hong Kong influenza did not start until October. It spread gradually, first to the larger and more crowded cities and then to smaller urban areas, but in a sporadic manner and without massive involvement of the rural areas. This was in marked contrast to an influenza B epidemic that affected the whole country during the same period-September-October 1968 to March 1969-and also in contrast to the 1957-58 Asian influenza epidemic. On 25 July 1968 information reached the Japanese Influenza Centre that an influenza epidemic with an unusually high attack rate was taking place in Hong Kong. On 31 July the present author went to Hong Kong at the request of the Japanese Govern- ment to obtain detailed information on the epidemic and to bring back strains of the virus causing the epidemic. He returned on 3 August with several isolates given to him by Dr W. K. Chang of the Hong Kong Influenza Centre, which were promptly submitted to antigenic analysis. Also on 25 July, an Israeli cargo vessel, the Tevirya, arrived from Hong Kong at the Japanese port of Nagoya with a number of the crew ill with influenza. Two strains of influenza virus were isolated on 1 August and forwarded to the Japanese Influenza Centre laboratories for analysis. On 12 August, the Japanese Influenza Centre announced that the isolates from the Hong Kong epidemic seemed to belong to a new antigenic variant, with an antigenic shift so considerable that it was highly improbable that the available vaccine, prepared from previously isolated A2 strains, would be effective against an epidemic caused by the new variant. Shortly afterwards, the Centre received essentially similar information from the World Influenza Centre, except for a very minor point of nomen- clature in that the World Influenza Centre considered the new strain to be a variant of A2, calling it 1 Chief, Department of Bacteriology, National Institute of Health, Tokyo, Japan. A2/Hong Kong, while the Japanese Influenza Centre inclined rather to the view that it should be called A3. The present paper describes further developments of the Hong Kong influenza epidemic in Japan. IMPORTATION OF HONG KONG INFLUENZA VIRUS INTO JAPAN It was not until early October 1968 that a true epidemic of Hong Kong influenza began in Japan, although more than a dozen importations of Hong Kong virus were reported in August and September and confirmed (mostly by virus isolation and in some instances serologically). These came not only from Hong Kong but also from other countries subse- quently involved in the epidemic but did not imme- diately lead to further spread of the infection. There were very few infections among contacts of influenza patients. For instance, although many of the crew of a fishing boat from Goto Islands, Nagasaki, in western Japan, returned in September from Taiwan with what was later virologically confirmed as Hong Kong influenza, the infection showed no tendency to spread to other people in the town or to the neighbouring villages. Also of interest are a few laboratory infections which occurred in vaccine-manufacturing labora- tories. On 13 August 1968 the Japanese Influenza Centre delivered a 3rd egg-passage virus, A2/Aichi/2/ 68 (Hong Kong isolate from the vessel Tevirya), to those laboratories as the strain for vaccine manu- facture. A number of laboratory infections, con- firmed by virus isolation, were reported from almost 2374 -33 - H. FUKUMI all the manufacturers. It was said that the infections took place only when virus of early passage was being used. A very few contact cases in families were reported, but the infections soon subsided and neither showed any tendency to spread further nor seem to have become sources for subsequent outbreaks. THE START OF THE HONG KONG EPIDEMIC IN JAPAN All reported cases and outbreaks of influenza-like disease were investigated (by virus isolation, sero- logical diagnosis or both) by central, prefectural or other laboratories but none was confirmed by laboratory methods as being due to Hong Kong virus until the beginning of October 1968, except for imported cases, laboratory infections and their immediate contacts. The first Hong Kong influenza outbreak occurred in Ryogoku Middle School in Tokyo, and was gradually followed by others, first in the Tokyo-Yokohama area and then in Osaka, Nagoya and other prefectures, almost all these outbreaks being in schools. Although there were no confirmed cases or out- breaks between August and October, it seems fairly certain that a smouldering seeding of virus had been occurring from imported cases because, as shown in Table 1, Hong Kong influenza outbreaks were reported in the earlier part of the epidemic chiefly in or near the ports of virus importation. This is in marked contrast to the B-type influenza epidemic which took place during the same period. Fig. 1 shows the general trend of influenza-like disease by week from May 1968 through April 1969. Since it is not based on laboratory examina- tions, it cannot indicate the virus type responsible, (although both B and Hong Kong types must be involved) or exclude non-influenzal respiratory FIG. 1 CASES OF INFLUENZA-LIKE DISEASE REPORTED IN JAPAN, MAY 1968 TO APRIL 1969,a AND NUMBERS OF SCHOOL CLASSES CLOSED b In Ir 0 0 In D z 100 000.~0o000 60 000 40 000 20 000 10000.- 8000.- 6 000 . 4000.- 2 000 1000 - Soo -600 . 400 - 200 100 -80 - 60 - 40 - 20 - 10 - a - 4 - * Nuumber of classes closed owing 0o influenza o Number of cases reported as influenza M691632 WEEKS IN 1968 WEEKS IN 1969 a The arrow indicates the time of the first outbreak of Hong Kong influenza (in Tokyo). b Schools were closed for the summer holidays during weeks 30-36. 354 355HONG KONG INFLUENZA IN JAPAN TABLE 1 CONFIRMED INFLUENZA OUTBREAKS IN JAPAN, SEPTEMBER 1968 TO MARCH 1969 Hong Kong influenza Type B influenza Prefec- ture a Oct. Nov. Dec. Jan. Feb. March Sept. Oct. Nov. Dec. Jan. Feb. 1968 1968 1968 1969 1969 1969 1968 1968 1968 1968 1969 1969 ______ .____ _____ _____ ___ _____ _____ _____ _____ _____ _____ _____ _____ Hokkaido Aomori Iwate Akita Miyagi Gunma I baragi Tochigi Saitama Chiba Tokyo Kanagawa Yaman- ashi Shizuoka Aichi Gifu Miye Fukui Toyama Shiga Kyoto b Osaka Waka- yama Hiroshima Tottori Ehime Toku- shima Fukuoka Saga Kago- shima Kuma- moto Nagasaki 2 5 2 4 2 3 2 2 4 2 1 2 3 5 2 5 3 1 3 3 4 6 4 8 3 12 6 5 5 3 9 2 3 6 4 10 8 2 2 3 1 3 4 2 2 1 2 2 1 2 2 2 4 1 2 2 3 1 2 I 4 1 5 3 4 6 2 5 1 1 8 3 2 5 3 3 2 7 2 5 7 1 8 4 2 2 7 2 1 3 a Prefectures in italics are ports of virus importation. b There was also an outbreak of type B influenza in Kyoto in May 1968. 1 H. FUKUMI diseases. Nevertheless, it does serve to outline the picture and shows clearly that there was a main epidemic period in January to February 1969, preceded by a wave of moderate size, whose start almost corresponds with the first Hong Kong influenza outbreak in Ryogoku Middle School. Fig. 1 also gives the curve for the number of classes closed owing to influenza-like illnesses; the curve (not shown) for the number of schools closed for the same reason closely follows that for the classes. PREVALENCE OF HONG KONG INFLUENZA IN JAPAN Even after the Hong Kong influenza epidemic had started, its spread was not as rapid as might be expected when a new antigenic variant appears and it can hardly be said to have developed typically epidemic features until January 1969. And then the epidemic still seemed reluctant to increase its speed of spread or to infiltrate further from urban into rural areas, a feature that was not expected from experience with the Asian influenza epidemic in 1957. Prior to the Hong Kong influenza epidemic, only a small fraction of the population of Japan had any Hong Kong antibody and then at titres as low as 1: 16 to 1: 32 or at most 1: 64 (see, for instance, Table 2). Table 2 is presented as an example, taken from findings in Japanese Self-Defence Forces camps in many parts of the country, to show to what extent the Hong Kong influenza epidemic affected com- munities. The results (not tabulated here) of a number of serological investigations in other popula- tion groups conducted immediately before the epidemic suggest that the antibody patterns in these camps are representative of the country as a whole. Sera were drawn twice in the camps-in October 1968, just before the epidemic, and in April 1969, after the epidemic had subsided. Taking a 4-fold antibody rise as the criterion, 495 out of 1325 soldiers are considered to have been infected with Hong Kong influenza, an infection rate of 37 %. The infection rates for the various camps have been plotted on a map of Japan in Fig. 2, the percentages of those with a 4-fold antibody rise being shown within double parentheses-e.g., Fukuoka ((25 %)). Where paired October/April sera were not available and serum samples were drawn only in the period TABLE 2 SHIFT OF HONG KONG ANTIBODY TITRES IN RANDOMLY SAMPLED INDIVIDUALS IN SELF-DEFENCE FORCES CAMPS IN JAPAN, OCTOBER 1968 TO APRIL 1969 Hi antibody titre in April 1969 Total <1: 16 1:16 1:32 1:64 1:128 1:256 1:512 1:1024 1:2 048 1:2 048 | 1:1024 D 1:512 -o 1:256 1 u 0 1:128 1 C o. 1:64 3 2 1 15 ~0 D2 1: 32 6 4 14 1 1 3 1 39 X1 :16 13 a 15/ 4 2 2 6 50 <1::16 70 71 97 133 71 25 10 1219 Total 729 57 89 85 107 139 80 28 11 1 325 4-fold or greater anti- body rise No. 60 75 104 138 79 28 11 495 67 85 97 99 99 100 100 356 HONG KONG INFLUENZA IN JAPAN 357 FIG. 2 HONG KONG INFLUENZA INFECTION RATES (CRUDE AND CORRECTED) a IN JAPANESE SELF-DEFENCE FORCES CAMPS Mihoro (I15%)123%l Nayoro (22%) [38%) Asahikawa ((21%)) Higashichitose ((47%)) / Sapporo ((38%)) / Hirosaki(1 0%)[2 1%l bihiro ((43%)) Akita (9%) (17%) Z inmachi ((2 6%))) Aomori ((33%)) Takada (17%) (38%) Hachinohe (41%)[34%) Sohmagahara (4%) (19%) Matsumoto (31%)[43%1 Kanazawa (23%)[40%1 Sendai (26%)[37%) \ukushima(17%)[23%) \ (J\ d tX1 Koriyama(1O0%)(13%l Yon,ago(l 1%)(20%l \ \g\ f ,7_ Utsunomiya(28%)[46%3\ \ % 14/ > ~~~~~~~Shinmachi ((45%))Kaidaichi((38%)) JIh\.l,Cc-'5A Yamaguchi(17%)(28%) \? , Nerimaa((76%)) Fukuoka (( 2 5% ))\\ j\ ohm((4%) Ai,,oiwa)17%)(27%] ~~~~~~~~~~~Toshim,a(48%))Ainoura (1 7%)[27%| g'ui(6)tS Fuji )46%)(57%] Toyokawa (29%)[42%I Nagoya((35%)) WHO91631 Ohtsu (( 56%)) Senzo (21%)(34%l Himeji)(12%)(28%1 Z entsuji (30%)t38%] Matsuyama( 18%)[33%) al \ \ Beppu t29%)l9¶% Kumamoto ((1 3%)) Kokubu ()2%) 4%l a Paired sera were drawn in October 1963 and April 1969; single sera from early May to early July 1969. Figures within single parentheses-e.g. (17%)-are percentages of persons with antibody titres >1 :128; figures within square brackets-e.g.[28%]-are the corrected values (see text). Figures within double parentheses-e.g., ((23%))-are percentages of persons showing a 4-fold or greater antibody rise in the paired sera. May to July 1969, the infection rate was calculated on the assumption that an antibody titre > 1: 128 represented infection since there was no one with so high a titre in October 1968 (see Table 2); this rate is shown in Fig. 2 within single parentheses-e.g., Yamaguchi (17 %). However, among those with antibody titres of 1: 32 or 1: 64 after the epidemic, some were found to have been infected on the basis of a 4-fold antibody rise. A correction factor was therefore calculated to take this into account and applied to the values within double parentheses in Fig. 2; these corrected values are shown within square brackets-e.g., Yamaguchi [28 %]. It will be seen from the figure that the infection rate in the camps was variable, ranging from [57%] to [4.0%] during the epidemic, and it was considerably lower than in the Asian influenza epidemic of 1957-58 (Fukumi, 1959), particularly in the west and south (Kokubu, Kumamoto, Ainoura, Fukuoka, Yama- guchi, Yonago), the north (Akita, Hirosaki, Mihoro) and some other areas (Koriyama, Fukushima, Sohma- gahara). It seems to run parallel with the prevalence in the civil population. AN EXAMPLE OF HONG KONG INFLUENZA IN THE EARLY EPIDEMIC PERIOD IN SCHOOLCHILDREN The epidemic situation is typified by the records of the Ohmukou Primary School, located in a fairly crowded part of Tokyo and involved in the Hong Kong influenza epidemic at the end of October 1968. The numbers of children with influenza-like illness in each class of the 4th, 5th and 6th grades are given in Table 3 by 5-day or 6-day period. It will be seen 358 H. FUKUMI TABLE 3 NUMBERS OF CHILDREN WITH INFLUENZA-LIKE DISEASE IN OHMUKOU PRIMARY SCHOOL, BY SCHOOL GRADE AND CLASS, OCTOBER-NOVEMBER 1968 I_______i ______- Date in October Date in November Total Grade Class affected/Total 1-5 6-10 11-15 16-20 21-25 26-31 1-5 6-10 in class 1 0 1 2 1 1 4 1 0 10/37 2 2 1 1 0 2 1 2 0 9137 4th 3 3 1 0 1 2 6 5 0 18137 4 0 3 0 1 0 3 2 0 9/38 1 0 3 0 3 2 1 8 5 22/48 5th 2 2 1 1 0 2 3 5 3 17/48 3 0 2 2 0 0 4 13 2 23/47 1 1 1 2 1 2 6 4 1 18/49 6th 2 2 2 1 0 0 4 8 3 20/48 3 1 0 2 3 14 17 1 0 38/48 that there was a remarkable accumulation of there should have been such a difference between patients in class 3 of the 6th grade at the end of two classes in the same school or why the epidemic October but not in the other classes except for a did not promptly spread from class 3 of the 6th slight clumping of cases in class 2 of the 6th grade grade to the other classes cannot be answered from and class 3 of the 5th grade in early November. the serological or epidemiological data. However, On 12 November sera were collected from the events of this sort were a general feature of the children in classes 2 and 3 of the 6th grade and epidemic, especially at the start. examined for their Hong Kong antibody titres (Table 4). STUDY IN TWO TOKYO SANATORIA Of the 48 children in class 3, 33 had antibody titres > 1 : 128. If these are taken as indicating In another study, two lots of sera were collected infection, there is a 69% infection rate. However, in two sanatoria in the outskirts of Tokyo and tested if a titre of 1: 64 is considered to represent possible for Hong Kong and B-type influenza antibody. infection, then the rate rises to 75 %. In class 2, on In Fuchu sanatorium they were drawn from 114 the other hand, only 3 had antibody titres 2 1: 128; patients in mid-February and mid-April; and in and even if those with a titre of 1: 64 are taken into Nakano sanatorium from 484 patients and 109 consideration, the infection rate is only 13 %. Why sanatorium personnel in mid-February and at the TABLE 4 FREQUENCY DISTRIBUTION OF HONG KONG ANTIBODY TITRES IN A NON-EPIDEMIC AND AN EPIDEMIC CLASS IN OHMUKOU PRIMARY SCHOOL Hong Kong antibody titre Class [ <1:16 1:16 ] 1:32 [ 1:64 J 1:128 1:256 1:512 1:1024 >1:2048 Grade 6, class 2(non-epidemic) 40 0 1 3 0 1 1 1 0 Grade 6, class 3 (epidemic) 13 0 0 2 2 7 10 9 5 HONG KONG INFLUENZA IN JAPAN 359 end of May. The results showed that, while the Hong Kong and B viruses were both already present in February, there were further infections with both types between the first and second serum collections. By the time the epidemic subsided, the Hong Kong virus infection rates amounted to 19% in Fuchu sanatorium (on the basis of a 4-fold antibody titre rise or of a > 1:128 titre)_or to 29% (taking a 1: 64 titre as the criterion). The corresponding figures in Nakano sanatorium were 34% or 40%. These figures indicate that some 30 %-40% of people became infected by the end of the epidemic in the Tokyo area, or at any rate in such groups as the inmates of sanatoria, which are relatively closed communities but with some contact with the general population. REFERENCES Fukumi, H. (1959) Bull. Wld Hlth Org., 20, 187-198