epidemiology of lobar pneumonia* |

Epidemiology of Lobar Pneumonia* |WILSON G. SMILLIE, M.D., F.A. P. H. A., AND

FREDERICK S. LEEDER, M.D.Professor of Public Health Administration, and Research Epidemiologist,

School of Public Health, Harvard University, Boston, Mass.

N a preliminary study one of us 1 de-termined the prevalence of various

types of pneumococci in the nasophar-ynges of persons who had been inintimate contact with cases of lobarpneumonia. Nasopharyngeal culturesof over 1,000 persons (contacts andcontrols) indicated that Types I and IIpneumococci were much more prevalentin the nasopharynges of immediatefamily contacts of cases of lobar pneu-monia due to the homologous organismsthan in the population at large. Thissuggested the advisability of furtherstudy in an attempt to determine thesignificance of this finding. Does thecase of lobar pneumonia due to Type Ior II pneumococcus infect his immediatecontacts; or do these types invade thenasopharynges of a family, establishthemselves and ultimately producepneumonia in one unfortunate member?An answer to at least a part of thisquestion might be reached by first tak-ing cultures of the family contacts ofan individual who had developed lobarpneumonia. One would then removethat person to a new uninfected environ-ment, thus exposing a new, presumablysusceptible and uninfected group ofcontacts, and make a study of this newgroup. Our approach to the problemwas to restrict ourselves to a study ofthe contacts of cases of lobar pneu-

* Read before the Epidemiology Section of theAmerican Public Health Association at the Sixty-second Annual Meeting in Indianapolis, Ind., October9, 1933.

monia which were due to Type I or IIpneumococcus and that were taken tothe wards of general hospitals early inthe course of the disease. We com-pared the nasopharyngeal content of thehome contacts of each case with thenasopharyngeal content of hospital con-tacts of the same case.The hospitals of Metropolitan Boston

cooperated in the study and informed usimmediately by telephone whenever acase of lobar pneumonia due to eitherType I or Type II pneumococcus wasadmitted to their wards. Our procedurewas to culture immediately the new en-vironment of the patient; namely, thehospital patients adjoining the new case(if the case was in a large ward), orall the contacts, if the new case wasplaced in a small 4- or 5-bed ward.This enabled us to establish the naso-pharyngeal flora present in the newenvironment at the time of the intro-duction of a source of Type I or IIpneumococcus. These contacts werethen re-cultured at intervals as long asthe contact remained unbroken, and aslong as the pneumonia patient continuedto exhibit Type I or II pneumococci inhis sputum.The environment from which the

patient had come was studied concur-rently. The family from which the casehad been removed was visited, andnasopharyngeal cultures were takenfrom as many of the immediate familycontacts as possible.As a further index of the extent to

which Type I or II pneumococcus[129]

AMERICAN JOURNAL OF PUBLIC HEALTH

TABLE INASOP:iARYNGEAL CULTURES ON HOSPITAL PATIENT AND FAMILY CONTACTS OF CASES OF LOBAR

PNEUMONIA DUE TO TYPE I PNEUMOCOCCI

Per CentPositive Positive Positive

Contact and not and andCases Contacts Days Negative Homologous Homologous Homologou.;

HospitalContactsFamilyContacts

20 61 426 38 50

17 62 153 13 34

strains are transmitted from acute casesto their contacts, house officers, nurses,and orderlies working on the wards inwhich lobar pneumonia caused by thesetypes of pneumococcus were beingtreated were cultured. These cultureswere taken before exposure to the case,during exposure, and at intervals afterthe exposure had ended.

Several families in which the inci-dence of homologous carriers was highwere cultured at regular intervals todiscover the length of time a " healthy "carrier was likely to carry a Type I orII pneumococcus.

Various factors of the environmentwhich might influence the transfer ofpneumococci from patient to contact,such as overcrowding, low economic orsocial status, were noted. Similarly,variations in the individual contacts,such as age, the existence of respiratorydisease, or the occurrence of acuterespiratory infection, chilling of thebody by exposure, etc., received atten-tion.The pneumococci were identified by

1.6

15 24.2

standard mouse passage and tubeagglutination technic as recommendedby Cooper 2 whose specific sera from Ito XXXII, inclusive, were employed.Our Type XX serum cross-agglutinatedwith so many types of pneumococci thatit was used only occasionally, whichprobably accounts for our finding thistype but once. The virulence to amouse of each strain of pneumococcusthat was isolated from a contact wasdetermined. The technic and personnelfor both this study and the one pre-viously reported by Smillie 1 were thesame, so that the results of the twoare comparable.The general study was continued

through a 9 months period, Novemberto July, inclusive.

FAMILY, PATIENT, AND HOSPITAL STAFFCONTACTS OF LOBAR PNEUMONIA DUE

TO TYPE .I OR II PNEUMOCOCCUSThe contacts of 64 cases were studied.

Of these 64 cases, Type I was the in-fecting organism 42 times, and Type II,22 times.

BLE IINASOPHARYNGEAL CULTURES ON HOSPITAL PATIENT AND FAMILY CONTACTS OF CASES OF LOBAR

PNEUMONIA DUE TO TYPE II PNEUMOCOCCI

PositiveContact and not

Cases Contacts Days Negative Homologous

11 33 281 25 25

Per CentPositive Positiveand and

Homologous Homologous

3.0

10 37 134 9 22

HospitalContactsFamilyContacts

130

6 16.2

EPIDEMIOLOGY OF LOBAR PNEUMONIA 1

DIAGRAM 1

PNEUMOCOCCI ISOIATED FROM THE NASOPHYNGES OF FOUR HOUSE OICERS AND THRNURSES IN CHARGE OF A WARD IN 'WHICH A CASE OF TYPE I LOBAR PNEUMlONIA WAS TREATEDHospital Staff Contacts

1H.O.

11.O.

1H.O.

H.O.

N.

N.

N.Not NotTaken Taken

Jan. 24 Jan. 30

*Ind(icates the presence of a cold1.0. indicates House OfficerN. indicatas Nurse.

Family contacts were considered onlywhen they had been in contact withthe patient within 7 days; and hospitalpatients were cultured only if theyoccupied contiguous beds in a largeward or were in a small 4- or 5-bedward with the case. Thus, all contactscould be called intimate from the pointof view of ease of exchange of naso-pharyngeal flora by droplets.The distribution of cases and contacts

was: 61 patient contacts of 20 Type Icases; 62 family contacts of 17 Type Icases; 33 patient contacts of 11 TypeII cases; 37 family contacts of 10 Type

at the time the culture was taken.

II cases; and 71 hospital staff contactsof 5 Type I and 1 Type II cases.A summary of the results of the

study of these 5 groups is shown inTables I, II3 and III. Diagram Iillustrates the details of a typical studyon hospital staff contacts.The striking thing about the tables

is the high percentage of homologouscarriers among the family contacts, ascompared to the low percentage amonghospital staff or patient contacts. Thisdifference is further emphasized whenwe compare the number of contact daysfor family contacts with the number of

131


TABLE III

NASOPHARYNGEAL CULTURES ON HosPiTAL PATIENTS; HosPiTAL STAFF AND FAMILY CONTACTSoF CASES oF LOBAR PNEUMONIA DUE TO TYPE I OR II PNEUMOCOCCI

ContactCases Contacts Days

Per CentPositive Positive Positiveand not and and

Negative Homologous Homologous Homologous

31 94 707 63 75

6 71 610 42 2 7

27 99 287 22 56 21 21.2

contact days for hospital staff andpatient contacts. Sixty-two family con-tacts of 17 Type I cases were exposedfor 153 days and yielded 15, or 24.2per cent, homologous carriers, whereas61 hospital patient contacts of 20 TypeI cases were exposed for 426 days andyielded only 1, or 1.6 per cent,homologous carriers. A similar rela-tionship exists for contacts of Type IIcases. Thirty-seven family contacts of10 Type II cases yielded 6, or 16.2 per

cent, homologous carriers for 134 daysexposure, whereas 33 hospital patientcontacts of 11 Type II cases yieldedonly 1, or 3.0 per cent, homologouscarriers after 281 contact days. Simi-larly, 71 hospital staff contacts of 5 TypeI and 1 Type II cases yielded only 2, or2.8 per cent, homologous carriers after610 days of exposure.Another striking thing in relation to

family contacts should be emphasized,namely, that one is just as likely to findhomologous carriers among the familycontacts of cases of Type I or II lobarpneumonia on the first day of exposureas after a week of exposure. That is,the number of contact days followingthe onset of the lobar pneumonia doesnot appear to affect the carrier rate inthe home. This, coupled with the lowrate of spread in hospitals, suggeststhat there must be some additionalfactor which determines whether or

not a Type I or II pneumococcus be-comes established in the nasopharynxof persons that are exposed to cases oflobar pneumonia due to these particu-lar strains.

In order to study this phenomenonfurther, we determined to introduce a

known healthy carrier of Type I pneu-mococcus into a group of normal per-sons that were free from pneumococciof this particular type, and then tomeasure the spread of the organism inthe new, uninfected environment, in-cident with the culturing of a groupof controls in the Wrentham StateSchool for Mental Defectives, wediscovered that an inmate, E. H., age28, harbored a strain of virulent TypeI pneumococcus in her nasopharynx.E. H. gave no history of previouspneumonia, nor of contact with a case.E. H. was in Building F, housing a

group of slightly mentally retardedfemales, averaging about 25 years ofage, so that the group was fairlysimilar to one of normal adults. Onthe campus was another building-B-with a comparable group of women ata slightly higher age level, averagingabout 30 years of age. Each buildingwas a unit, with little contact frombuilding to building, or with the out-side.The plan followed was to culture

first the occupants of Building F, in

HospitalPatientContactsHospitalStaffContactsFamilyContacts

2 2.1

2 2.8

132

EPIDEMIOLOGY OF LOBAR PNEUMONIA

which E. H. had been a patient forseveral years. We also cultured theoccupants of Building B in order toestablish the normal nasopharyngealflora of the persons in the two build-ings. We then transferred E. H. toBuilding B and cultured, at frequentintervals, her new contacts in this newenvironment.

Building F contained 116 girls. Theywere about equally divided between 3wards in which the beds were arrangedas elsewhere in the institution; thehead of one touching the foot of thenext one, and the rows separated by anarrow aisle just wide enough to allowa person to walk between the rows.The night time contact, therefore, could-be consdered closer than would befound either in the home or in a gen-

eral hospital. Building B housed 97girls under exactly the same livnng con-ditions, so that the degree of contactin the two buildings was comparable.The two buildings gave a total of

213 persons. Of these 103 (48.3 percent) harbored pneumococci of varioustypes, and 110 (51.7 per cent) did not.These percentages are compatible withthe figures obtained by other workersfor the normal population. Gundel,3 in1,114 cultures of school children, found734 strains of pneumococci, andSmillie,' in 458 cultures of a cross-section of the normal population, found212 strains. In our series of culturesfrom Building B and F, all of the 32Cooper types except II, XXII, XXIII,XXVI, XXVII, XXIX, XXXI, andXXXII were recovered. Types III,

TABLE IVTYPES OF PNEUMOCOCCI PRESENT IN THE NASOPHARYNGES OF CONTACTS WITH A TYPE I

PNEUMOCOCCus HEALTHY CARRIER BEFORE AND AFTER THE INTRODUCTIONOF THE INFECTED PERSON INTO THE DORMITORY

Type of Pneumococcus

July

Case 11 17 18 19 21 24 27 28No.

Type I Carrier _ - - - * I I I I

1 - Neg. - * VI Neg. VI Neg.2 - Neg. - * Neg. VI VI VI

Contacts in con- 3 III _ * III Neg. IV Neg.tiguous beds 4 _ - VI * VI XXVI VI VIand at samedining table 5 VIII - _ * viii ViII viii ViII

6 - - Neg. * III VI VI VI

7 Neg. - - * _ Neg. _ VIIContacts at same

dining table, 8 Neg. - - * II IIbut not in con-tiguous beds 9 VII - - * - VIII VIII VIII

10 VII - - * - XI, XVI - XI, XVI

Contacts in con-tiguous beds, 11 - - Neg. * III - Neg. Neg.but not at samedining table* Type I carrier introduced on this date

133


VII, X, and XVIII were found mostfrequently, but no single type pre-dominated. Type I was not found inBuilding B. In Building F, where thecarrier was an inmate, 2 other girlsyielded Type I pneumococcus on oneoccasion. It was impossible to obtainthis organism from these 2 inmates onrepeat cultures. Only E. H. was con-tinually positive for the virulent TypeI pneumococcus.The culturing of the two buildings

was started on June 28, and was com-pleted on July 19, on which date E. H.was transferred from Building F toBuilding B, and her new contacts werethen cultured at intervals. Our pro-cedure by days follows:

July 21-Contacts in contiguous beds-Building B

July 21-Previous contacts in contagiousbeds-Building F

July 24-Contacts at same dining table-Building B

July 26-Contacts in same ward, but notin contiguous beds-Building B

July 26-Contacts in same building, butnot in same ward-Building B

July 26-Previous contacts not in sameward-Building F

July 27-Contacts in same building, but notin same ward-Building B


July 28-Contacts at same dining table-Building B


It will be noted that we concen-trated our efforts on contacts occupyingbeds contiguous to E. H. and to con-tacts eating at the same dining table.Many of the people in the contiguousbeds also used the same dining table, sothat some had more contact than others.As a check, cultures were also takenfrom time to time in the other wards inBuilding B and in Building F fromwhich E. H. had come. None of thesecontrol cultures were positive for TypeI pneumococcus.The results of one group of serial

cultures of intimate contacts of E. H.in her new environment are shown inTable IV. It is obvious from this tablethat E. H. did not infect her new con-tacts with Type I pneumococcus,though she continued to carry thisstrain throughout her stay in BuildingB.

Analysis of all the data, consisting ofapproximately 125 nasopharyngeal cul-tures of immediate contacts, reveals thatthe introduction of an individualharboring Type I pneumococci in thenasopharynx into a group of peoplehaving continued close contact with thecarrier was without effect. This state.ment, however, is modified by the lengthof contact, which was fairly short, July19-28; and by the time of the year,July-a month during which the inci-dence of pneumococci in the populationis low, and in which little lobar pneu-monia occurs. Additional importantdata would be secured if one repeatedthe Wrentham experiment during thewinter months, or during an outbreak ofupper respiratory disease, such as thecommon cold.We have some evidence that Type I

pneumococci, under certain conditions,may be transmitted through contact inthe hospital. In one instance of closecontact of a hospital staff member withan infected person, it would appear thata strain of Type I pneumococcus wastransmitted from patient to contact andthat lobar pneumonia due to Type Ipneumococcus had resulted.On March 9, a girl 2X2 years old,

suffering from lobar pneumonia, com-plicated by an empyema, was hos-pitalized, and Type I pneumococci wereisolated from the nasopharynx, andfrom the pleural fluid. We made serialnasopharyngeal cultures on 7 doctorsand 3 nurses who came in contact withthis child. On March 20, Dr. A., whowas in charge of aspirating pus from thepleural cavity, developed lobar pneu-monia due to Type I pneumococcus.

134


Dr. B. assumed care of the child, andon March 25 a nasopharyngeal cultureof his throat revealed Type I pneumo-coccus, but he did not develop pneu-monia. The remaining 5 doctors and 3nurses were cultured repeatedly untilApril 11, on which date the case ceasedto exhibit Type I organisms. None ofthem became positive for Type Ipneumococcus.

Since this child was the only knownsource of Type I pneumococcus in thehospital, it would seem reasonable toassume that this patient was responsiblefor the spread of the infection to the2 doctors, the spread probably beinginfluenced by the intimate contact as-

sociated with the aspiration of thepleural cavity of a small child.

Several families in which there oc-curred a high percentage of contactsthat carried Type I or II pneumococciwere studied with special care.

Family A-The mother of Family Abecame ill on January 14 with lobarpneumonia. The infecting organismwas Type II pneumococcus. On thefollowing day a small daughter de-veloped lobar pneumonia, also causedby a Type II pneumococcus. Bothpatients were removed to a local hos-pital on January 15. On January 17,the 6 remaining members of the family,1 man, 2 boys, and 3 girls, were cul.-

DIAGRAM 2

PNEDJOCCI ISOLATEI PM TE NASOPHANGES OF SIX ERSOF A FAM= IN ISICH TWO CASES OF T!PE II LOBAR PNEMNIA OCCURRED

Jan-17 Feb.1 Feb.17 Mar.6

*Idiotes the presence of a cold at the ti the culture was taken.

135


DIAGRAM 3

PNEUMOCOCCI ISOAT ED 7O E NASOPBAR SOF FOUR PATIENiTSOCCUPXING THE SAME FIVE-BED WARD AS A CASE OF TrPE I LOBAR PNEUMONIA

Pait Contacts

Case hospitalisedJanuary 15. Dis-charged February 1.

*Indicates that no culture was obtained.

tured. Type II pneumococci were re-covered from the nasopharynges of 4 ofthese contacts. Two that carried TypeII pneumococci had bad colds at thetime the cultures were taken, and 2had not had bad colds for more than amonth.

These 6 contacts were recultured atabout 2-week intervals until 5 serialcultures had been obtained (DiagramII). It will be noted from the diagramthat 1 continued positive for Type IIpneumococcus throughout; 1 yieldedType II pneumococcus only on the firstculture; 1 yielded Type II on the sec-ond culture, became negative to TypeII, and finally exhibited a Type III;

and 1 yielded a Type II on the secondand last cultures, having been negativefor Type II in the meantime.The 2 patients who had been in the

hospital with pneumonia returned homeon February 1, free from Type IIpneumococcus. The mother remainedconsistently negative, but her daughterbecame positive for Type II pneumo-coccus again, after being negative forthis strain on two previous cultures. Itis interesting to note that during the2½ months that the family was studiedit was never free from a source of TypeII pneumococcus, and that the 2 per-sons who became negative to Type IIpneumococcus and later positive both

136


gave a history of an acute cold at thetime the Type II pneumococcus wasrecovered.The mother of Family A was hos-

pitalized in a 5-bed ward and serialcultures were taken on the 4 patientsin her new environment. These havingon the average at least 4 days close con-tact with her, each failed to yield aType II pneumococcus (Diagram 3).

Family B-On January 23, themother of Family B developed lobarpneumonia due to Type I pneumo-coccus. On January 27, 4 days afterthe patient had been removed to a hos-pital, nasopharyngeal cultures weretaken on 10 home contacts, 6 of whomwere found to be carrying Type I pneu-mococci. In this particular family, 9-of the 10 contacts, which included all6 of the homologous carriers, hadsevere colds at the time the cultureswere taken (Diagram 4). This familywas not followed at short intervals, but

when recultured April 17, nearly 3months later, none of the contactsyielded Type I pneumococci. Theoriginal case, however, who had re-covered and returned home February16, was still positive for Type I pneu-mococcus on April 17.The mother of Family B was hos-

pitalized in a 5-bed ward and serialcultures were taken on the 4 patients inher new environment. As with FamilyA, the 4 patients, having at least 4days close contact with her, each failedto yield a Type I pneumococcus.Family C-In April, 1932, a 13 year

old boy in Family C developed lobarpneumonia due to Type II pneumo-coccus. Five family contacts that werecultured at the time yielded 4 type IIcarriers. This family was not followedat frequent intervals, but the hospitalrecords show that the child developed achronic empyema, from which Type IIpneumococci were recovered, and which

DIAGRAM 4

PNEMOCOCCI IS FOM THE NASOPHARYNGES OF TEN M BERSOF A FAMI IN WHCH ONE CASE OF TYPE I LOBAR PNEMOIA OCCURRBD

The case was hospi-talized Jaluel7 20and returned homeFebruaery 16.

-o

*Indicates the presence of a cold at the time the culture was taken.

137

138 AMERICAN JOURNAL OF PUBLIC HEALTH

continued to discharge for severalmonths after the child returned home.The family was recultured in April,1933, nearly a year later, and Type IIpneumococci were recovered from thenasopharynges of the original case and3 of the 4 original carriers.

THE RELATIONSHIP OF ACUTE RESPIRA-TORY DISEASE IN CONTACTS OF TYPE

I OR II LOBAR PNEUMONIA TO PREVA-LENCE OF PNEUMOCOCCI IN THE

NASOPHARYNXIt has been suggested that a person

with a cold is more likely to harborpneumococci than one without a cold.Our data indicate, however, thatpneumococci were not significantlymore prevalent in contacts sufferingfrom colds than in those free fromcolds. A different picture was pre-sented when we considered only thegroup of contacts of Type I or II caseswhich harbored the homologous or-ganism. Of out group of 25 homo-logous carriers, 19, or 76 per cent, hadcolds at the time they were cultured.The numbers are small, but the figuresare significant, and suggest furtherstudy in this direction.Our data indicate that overcrowded

living conditions in the home, and alsolow economic status were not factors inthe establishment of carriers of TypeI and Type II pneumococci.

SUMMARY AND CONCLUSIONS1. Nasopharyngeal cultures from 264 con-

tacts of 64 cases of lobar pneumonia due to

Type I and Type II pneumococcus have beenstudied. The results indicate that 20± percent of the immediate family contacts ofthese patients harbored the homologousstrain in their nasopharynges. The hospitalcontacts of the same patients, however, wereseldom infected (about 2 per cent). Theseresults suggest that it is quite justifiable totreat cases of lobar pneumonia due to TypeI and Type II pneumococci in the open wardsof our general hospitals.

2. One experiment indicates that a healthycarrier of Type I pneumococcus does nottransmit this organism to immediate contacts,even though the individuals are living underovercrowded conditions.

3. Our evidence suggests that there is someadditional factor other than simple contactwhich determines the transfer of Type I orII pneumococci from a patient with lobarpneumonia to contacts. Nineteen of the 25family contacts of cases of lobar pneumoniadue to Type I or Type II pneumococci thatbecame homologous carriers were sufferingfrom acute colds at the time the cultureswere taken. Positive cultures were found asfrequently on the first day of exposure as,after a week. This evidence suggests thatfamily epidemics of colds may be a factorwhich determines the transfer and establish-ment of Type I and Type II pneumococcifrom the infected to the uninfected.

4. Carriers of Type I and Type II pneumo-cocci, when once established, may continueas carriers of these strains for a considerableperiod of time without giving rise to lobarpneumonia in the carrier or his contacts andwithout producing a second group of carriers.

REFERENCES1. Smillie, W. G. J.A.M.A. In press.2. Cooper, G., et al. J. Exper. Med., 49:461,.

1929; 55:531, 1932.3. Gundel, M. Ztschr. f. Hyg. u. Infektionskr.,

V, 114:659-677, 1933.

epidemiology of lobar pneumonia* |

Documents