1980;65;258PediatricsMichael P. Sherman, Boyd W. Goetzman, Charles E. Ahlfors and Richard P. Wennberg

PneumoniaTracheal Aspiration and Its Clinical Correlates in the Diagnosis of Congenital

  

  http://pediatrics.aappublications.org/content/65/2/258

the World Wide Web at: The online version of this article, along with updated information and services, is located on

 

ISSN: 0031-4005. Online ISSN: 1098-4275.PrintIllinois, 60007. Copyright © 1980 by the American Academy of Pediatrics. All rights reserved.

by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village,it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication,

by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

mington and Klein8@'7@ suggested that direct laryngoscopy and tracheal aspiration (TA) be used toidentify bacterial etiologies in neonatal pneumonitis, but th€ydid not document the efficacy of thisapproach. This report evaluates the use of directTA for early diagnosis of congenital pneumonia,comparing data obtained by this technique withother clinical and laboratory findings consideredpredictive of neonatal infection.

MATERIAL AND METHODS

Patient Selection

The newborns studied were hospitalized in theIntensive Care Nurseries of the University of California, Davis Medical Center or the Kaiser Foundation Hospital and Medical Center, Sacramento,California. Infants were selected for TA if they metall of the following criteria: (1) age less than 8 hours;(2) either not intubated or intubated for less thanone-half hour duration; (3) one or more pennatalrisk factors for fetal infection; (4) early cardiorespiratory symptoms; and (5) an abnormal chest radiograph.

The perinatal factors associated with increasedfetal infection included premature ruptured membranes, premature labor, prolonged ruptured membranes or labor, maternal fever, suspected amniotitis, fetal heart rate abnormalities, prematurity,difficulty of delivery, or birth asphyxia. Cardiorespiratory abnormalities included symptoms of therespiratory distress syndrome, apnea, tachypnea,cyanosis, or hypotension. Radiologists independently read the radiographs without knowledge ofTA findings. Their interpretations included hyalinemembrane disease, retained lung fluid, and pulmonary infiltration.

Between January 1975 and January 1978, 320infants less than 8 hours of age met our study

ABSTRACT.Trachealaspirateswereobtainedfrom320newborns with respiratory distress and one or more perinatal risk factors for infection. Samples were obtainedbefore 8 hours of age, either by direct aspiration orimmediately following intubation. Twenty-five infantshad bacteria present in the aspirate smear. In each casea pure culture of the organism suspected by smear morphology was grown. The same organism was isolated fromblood in 14 of the 25 newborns suspected of havingpneumonia. The remaining 11 infants had clinical courses,depressed mature neutrophil counts, and elevated bandto total neutrophil ratios consistent with bacterial infection. Twenty-five infants without bacteria in the trachealaspirate smear were randomly selected as a comparisongroup. Three of these infants had positive blood cultures,and one of the three also grew the same organism fromthe tracheal aspirate. The chest radiograph was abnormalin all infants but did not discriminate patients with orwithout pneumonia. We conclude that examination of thetracheal aspirate obtained within the first 8 hours of ageis helpful in the early diagnosis of congenital pneumonia.Pediatrics 65:258—263,1980;tracheal aspiration, congenital pneumonia, band to total neutrophil ratio.

Approximately 20% of stillborns and neonatesdying before 72 hours of age have histologic evidence of pneumonia.―2The clinical diagnosis ofcongenital pneumonia is usually made on the basisof clinical events,3 hematologic profiles,4'5 radiographs,@8@'61 and bacteriologic cultures.8@'7' Specific diagnostic procedures, such as modified expectorated9 or transtracheal sputum analysis,'°― arenot applicable to newborns, and increased morbidity might result when diagnostic lung puncture'2 isused in serious neonatal respiratory disease. Re

Received for publication March 15, 1979;accepted June 4, 1979.Reprint requests to (R.P.W.), Department of Pediatrics, UCDavis Medical Center, 4301 X St, Sacramento, CA 95817.PEDIATRICS (ISSN 0031 4005). Copyright ©1980 by theAmerican Academy of Pediatrics.

258 PEDIATRICS Vol. 65 No. 2 February 1980

Tracheal Aspiration and Its Clinical Correlatesin the Diagnosis of Congenital Pneumonia

MichaelP. Sherman,MD, BoydW. Goetzman,MD, PhD,Charles E. Ahlfors, MD, and Richard P. Wennberg, MD

From The Kaiser Foundation Hospital, Sacramento, and The University of California,Davis Medical Center, Sacramento

by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

criteria. Twenty-five infants were designated suspect for congenital pneumonia because bacteriawere identified in the Gram stains of their TAs. Inno instance did we observe bacteria in the absenceof polymorphonuclear neutrophils (PMNs).Twenty-five infants were randomly selected fromthe larger, smear-negative group (no bacteria seen)for comparison with these suspect newborns.

Tracheal Aspiration and Specimens

Specimens were obtained by direct laryngoscopyand passage of an endotracheal tube or 1OF suctioncatheter into the trachea while avoiding oropharyngeal contamination. The suction catheter was attached to a sterile mucus trap (Figure). When onlya suction catheter was used to obtain a specimen,a 25-gauge scalp vein needle was inserted near thethumb holder and attached to a sterile, saline-filledsyringe. If no secretions were obtained, three 0.5-misaline lavages were performed to increase specimenyield. The infants were ventilated between suctionings to minimize hypoxemia from the procedure.

Tracheal aspirate smears were immediatelystained with Gram's solution and examined for bac

teria and PMNs. A Culturette swab (modifiedStuart's transport medium, Marion Scientific Corp,Rockford, IL) was saturated with the aspiratedmaterial for transport to the bacteriology laboratory. Cultures were performed using blood, chocolate, and MacConkey's agar plates, and thioglycolate broth. Specimens were not specifically processed for anaerobes, mycoplasma, or chlamydia.

Clinical and Laboratory Correlates

At the time of TA, all infants had vital signsrecorded, a complete blood and differential count,an arterial blood gas determination, and a bloodculture. Cerebrospinal fluid (CSF) was examined in

all smear-positive infants and in over half of thecomparison infants. Tracheal and gastric specimens

were obtained simultaneously in 14 infants. Noinfant received antibiotics prior to bacteriologicstudy.

Statistical Analysis

Group means were compared by Student's t test.The degrees of freedom were calculated by Welch'sformulas, which assume the two groups have tinequal variances.'3 Clinical observations were compared using x2 analysis with Yates' correction.'4

RESULTS

Patient Characteristics

Suspect and comparison groups had similar dinical characteristics (Table 1). There were no differences in mean gestational age, birth weight, Apgarscore, sex ratio, vital signs, arterial blood gases,need for resuscitation, or presence of early apnea(Table 1). The occurrence of perinatal events associated with increased fetal infection (Table 2) wassimilar in each study group, except that suspectinfants were subjected to a longer labor.

Maternal characteristics were similar in bothgroups with respect to age, gravity, parity, and peakintrapartum temperature. Five suspect and threecomparison mothers were suspected of having amnionitis, and three suspect and two comparisonmothers received antibiotics during labor.

The radiographic data is summarized in Table 3.Interpretations of hyaline membrane disease(HMD), retained lung fluid, and pulmonary infiltration occurred with a similar frequency in infantswith and without a positive TA smear. Some suspect infants having radiographs originally interpreted as hyaline membrane disease or retainedlung fluid later progressed to lobar infiltration. mitial radiographs revealed pleural effusions in foursuspect and two comparison infants.

Bacterial Investigation

All suspect infants cultured a single bacteriumfrom the TAs, and in each case a bacterium of thesame morphology was identified on the initial Gramstain of the TA (Table 4). One comparison infanthad a false-negative smear and subsequent positiveculture of the TA. Blood cultures were positive in14 of 25 suspect infants versus three of 25 comparison infants. All suspect infants had PMNs presenton the Gram stain of the TA. Eleven of 25 comparison infants had PMNs on their TAs. Two of these11 infants had positive blood cultures. Three suspect and two comparison infants had positive cerebrospinal cultures.

LSYRINGE(0.9%Saline)

Figure. Apparatus used for direct tracheal aspiration.

ARTICLES 259

by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

SuspectComparisonPremature

labor8/258/25Prematurerupture ofmembranes12/2511/25Duration

of ruptured membranes(hr)13.613.3Mean±SD25.212.6Duration

of labor(hr)11.2*6.5Mean±SD7.54.2Fetal

heart rate>1608/257/25Suspectedamnionitis5/253/25*

Significant difference, P = .01.

SuspectComparisonHyaline

membrane diseaseRetained lung fluidInfiltrate9

8812

85Total2525

SuspectPositiveSmearComparison

NegativeSmearPositive

trachealculture251Positivebloodculture143PositiveCSFculture32PMNs

present2511

TABLE 1. Characteristicsof the Study Population

Suspect

2,652±99536.3±3.6

187

Comparison

2,387 ± 719

35.4 ±4.0Birth weight (gm) (mean ±SD)Gestational age (wk) (mean ±SD)Sex

MaleFemale

OutbornApgar score

at 1 mm (mean ±SD)at 5 mm (mean ±SD)

Heart rate (beats/mm) (mean ±SD)Respiratory rate (breaths/mm) (mean

±SD)Systolic blood pressure (mm Hg) (mean

±SD)Rectal temperature (C) (mean ±SD)Arterial pH (mean ±SD)Apnea at <8 hr of ageInitial resuscitationSurvival

1510

9/25 8/25

5.7 ± 2.8

7.1 ± 2.5

157 ±1863 ±20

5.7 ± 2.5

7.8 ±1.5154 ±1661 ±16

62±9 58±13

36.9 ±0.47.32 ± 0.11

7/25

10/2519/25

37.1 ±0.77.32 ± 0.06

7/25

10/2522/25

* No significant differences between suspect and comparison groups were found for these

characteristics.

TABLE 2. Comparisonof Perinatal Risk Factors

TABLE 3. RadiographicFindings TABLE 4. Comparisonof MicroscopicandBacteriologic Findings

The specific bacteria cultured from the lowerrespiratory tract and blood are outlined in Table 5.The group B streptococcus (GBS) was the predominant tracheal and blood isolate in suspect infants.Two comparison infants grew GBS from the bloodand CSF, and the infant having the false-negativeTA-cultured Haemophilus influenzae from bothTA and blood.

Six suspect and eight comparison infants hadsimultaneous gastric and tracheal cultures. All sixsuspect infants had PMNs and pure Gram stainsand cultures for the same bacterium in the gastricand TA. Five of the eight comparison infants hadPMNs in their gastric smears. Two of these five

had mixed bacteria on smear and culture of thegastric aspirate. A single comparison infant hadGram-positive cocci without PMNs on gastric Gramstain and grew streptococci viridans.

Peripheral WBC Response

Eleven of the 25 suspect infants had positivetracheal but negative blood cultures. The polymorphonuclear cellular responses of all 50 infants wereanalyzed to help differentiate whether these 11infants were infected or simply colonized. Threegroupings of infants were compared; namely, (1)

260 TRACHEAL ASPIRATION AND CONGENITAL PNEUMONIA by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

TABLE 5. Bacteria IsolatedfromTrachealAspiratesTrachealBloodCSFSuspectGroup

Bstreptococci1472Hinfluenzae320Streptococcus

viridans200Ecoli210Listeria

monocytogenes221Staphylococcusaureus110Pseudomonasaeruginosa110Streptococcus

pneumoniae100ComparisonGroup

Bstreptococci022Hinfluenzae110

Corrected WBC countTrachea

(—)Blood(—)

14,860 ±7,330*Trachea

(+)Blood (+)

8,660 ±8,070P

Values

<.05Trachea

(+)Blood (—)

9,590 ±7,140P

Values

.10Totalneutrophil count9,550 ±7,9503,730 ±4,570<.014,880 ±4,330.05Absolute

mature neutrophils8,565 ±6,7002,260 ±2,500<.0013,500 ±3,470.01Absoluteimmature neutrophils1,380 ±2,3901,460 ±2,220<.501,360 ±1,350.50Band

to total neutrophil ratio0.13 ±0.190.39 ±0.23t<.050.42 ±0.30.02

infants having positive tracheal and blood cultures,

(2) infants with positive tracheal and negative blood

cultures, and (3) infants having negative trachealand blood cultures (Table 6). The total neutrophilcount and the absolute mature neutrophil countwere significantly lower in both the tracheal-positive groups. The absolute immature neutrophilcount did not differ among the three groups, andthis finding resulted in a significantly elevated bandto total neutrophil ratio in both tracheal-positivegroups. One TA and blood culture-positive infantwas excluded from ratio calculations because no

neutrophilic cells were present on the differentialanalysis.

Outcome

Six suspect and three comparison infants died.Eight of these died from apparent congenital infection. These included five infants with GBS infection: two of two comparison infants with septicemiaonly, two of seven suspect infants with both pneumonia and septicemia, and one suspect infant witha negative blood culture. Three suspect infants diedof infections due to Listeria, Pseudomonas, andHaemophilus, respectively.

Potential complications of TA include airwaytrauma, hypoxemia or cardiorespiratory depressionduring the procedure, and introduction of a pathogen into a diseased lower respiratory tract. Notrauma was apparent during or after the procedure.

Hypoxemia was brief during suctioning, and nocardiorespiratory deterioration occurred secondaryto the TA. One comparison infant who was ventilated from 16 hours of age for HMD grew Escherichia coli from a repeat TA and blood culture at72 hours of age. Tracheal and blood cultures were

negative at 1 hour of age, but E coli was isolatedfrom an ear swab at that time.

DISCUSSION

Direct tracheal aspiration performed within 8hours of age appears to be a specific, safe, and easilyapplied procedure for the early diagnosis of congenital pneumonia. The specificity of TA is predicated

on the assumption that pulmonary secretions arenormally sterile at birth and for several hours thereafter. This is true in vaginally delivered rabbitsexamined within 24 hours after birth,'5 althoughthere is some evidence that the respiratory tractmay be more rapidly colonized with pathogens inhuman neonates having lung disease.'6―7

Although microscopic evaluation of the trachealaspirate was remarkably specific and accurate inidentifying newborns with congenital pulmonaryinfection, the procedure did not delineate all infantshaving respiratory distress and sepsis. Perinatalhistory was more inclusive though far less specificin selecting infected infants. Except for a longerlabor in the TA-positive infants, clinical findingsdid not specifically separate infants having congen

TABLE 6. Hematologic Findings of the Study Population

* Mean ± SD.

t Oneinfanthad no neutrophiic forms,anda ratiocouldnot be calculated.

ARTICLES 261 by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

ithi pneumonia or other types of respiratory distressand sepsis. Initial resuscitation6―8 and earlyapnea,6―9findings previously correlated with earlyonset GBS disease, did not differ between our comparison and suspect infants having pneumonia ofvarying etiologies.

The difficulty in differentiating chest radiographsof early onset GBS disease from HMD has beenwell documented.6'7 In this study, GBS and otherbacteria were isolated from TAs of several patientswith roentgenographic interpretations of “¿�wetlung―or pulmonary infiltration as well as HMD. Conversely, negative TAs (and blood cultures) werealso associated with radiographic evidence of pulmonary infiltrate (Table 3). These observations emphasize the inability to diagnose congenital pneumonia using chest radiographs alone.

Gram stains and cultures of all gastric aspiratesfrom suspect infants yielded the same bacteriologicresults as the simultaneously obtained TA. Analysisof the gastric aspirate from the comparison infantsshowed that 60% had leukocytes and 37% had bacteria detected. Despite our small number of simultaneous gastric aspirates and TAs, our data suggestthat an increased number of noninfected infantswould be treated if PMNs or bacteria in gastricaspirates were the criteria for antimicrobial therapy. The study of Mims et al@°supports this observation, suggesting that gastric aspirates are no morehelpful than other high-risk perinatal events inselecting infants with infection. In addition, Vasonet al2' found that in cases of prolonged rupturedmembranes, gastric PMNs were of maternal originand were not indicative of fetal involvement.

Monroe et al4 have previously described the usefulness of the blood differential count in identifyingthose infants having early onset GBS infections andvariable radiographic findings. Boyle et al5 recentlydescribed similar hematologic proffles in early neonatal sepsis and respiratory distress caused by varying pathogens. We observed a depressed absolutemature neutrophil count and an elevated band tototal neutrophil ratio in infants having positivetracheal aspirates, regardless of whether the bloodculture was positive. These data strongly suggestthat tracheal culture-positive infants had significantpulmonary infection regardless of a positive bloodculture, and that hematologic changes help identifyinfants with congenital pneumonia.

Baker22 suggests that an unfavorable outcome forearly onset GBS disease may be predetermined byextensive intrauterine pneumonia. Although thisstudy was not designed to test the ability of the TAversus conventional diagnostic methods to influence survival, our 79% survival in early onset GBSdisease and 73% survival in other types of congenital

pneumonia compares favorably with otherstudies.5'6'8@2―'22'23Tracheal specimens were easilyobtained using routine respiratory techniques, andthe procedure appears to be safe, but further expenence with this method wifi determine if morbidityand mortality from this serious disease can be reduced through early identification and treatment.

In conclusion, early direct tracheal aspirationfacilitated the early diagnosis of congenital pulmonary infections. Gram stain morphology and tracheal isolates correlated well with the positive bloodcultures of infected infants. A depressed absolutemature neutrophil count and elevated band to totalneutrophil ratio were more frequently present ininfants having positive tracheal cultures, whetheror not a positive blood culture was obtained. Although the TA and hematologic findings may notidentify all cases of congenital pneumonia, they willprovide an early and specific diagnosis in the vastmajority of neonates with pulmonary infection. Finally, we were unable to identify any perinatal riskfactors or chest radiographic pattern which distinguished infants with or without pneumonia.

REFERENCES1. Fujihura T, Froehlich LA: Intrauterine pneumonia in rela

tion to birthweight and race. Am J Obstet Gynecol 97:81,1967

2. Naeye RL, Dellinger WS, Blanc WA: Fetal and maternalfeatures of antenatal bacterial infections. J Pediatr 79:733,1971

3. Avery ME, Fletcher BD: The Lung and Its Disorders in theNewborn Infant. Philadelphia, WB Saunders Co, 1974, p 175

4. Monroe B, Rosenfeld C, et al: The differential leukocytecount in the assessment and outcome of early onset neonatalgroup B streptococcal disease. J Pediatr 91:632, 1977

5. Boyle R, Chandler B, et al: Early identification of sepsis ininfants with respiratory distress. Pediatrics 62:744, 1978

6. Ablow R, Driscoll 5, Effman E, et al: A comparison of earlyonset group B streptococcal infection and the respiratorydistress syndrome of the newborn. N Engl J Med 294:65,1976

7. Lillien LD, Harris VY, Pildes RS: Significance of radiologicfindings in early-onset group B streptococcal infection. Pediatrics 60:360, 1977

8. Remington JS, Klein JO: Infectious Diseases of the Fetus

and Newborn Infant. Philadelphia, WB Saunders Co, 1976,pp 816, 817, 821

9. Barlett JG, Fmegold SM: Bacteriology of expectorated sputum with quantitative culture and wash technique comparedto transtracheal aspirates. Am Rev Respir Dis 117:1019,1978

10. Hahn HH, Beaty HN: Transtracheal aspiration in the evaluation of patients with pneumonia. Ann Intern Med 72:183,1970

1 1. Davidson M, Tempest B, Palmer DL: Bacteriologic diagnosis

of acute pneumonia: Comparison of sputum, transtrachealaspirates, and lung aspirates. JAMA 235:158, 1976

12. Klein JO: Diagnostic lung puncture in the pneumonias ofinfants and children. Pediatrics 44:486, 1969

13. Snedecor GW, Cochran WG: Statistical Methods, ed 2.Ames, IA, Iowa State University Press, 1967, p 115

14. Armitage P: Statistical Methods in Medical Research. NewYork, John Wiley & Sons mc, 1971, pp 134-135

15. Sherman M, Goldstein E, Lippert W, et al: Neonatal lung

262 TRACHEALASPIRATIONAND CONGENITALPNEUMONIA by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

defense mechanisms: A study of the alveolar macrophagesystem in neonatal rabbits. Am Rev Respir Dis 116:433, 1977

16. Spunt K, Leidy G, Redman W: Abnormal colonization ofneonates in an intensive care unit: Means of identifyingneonates at risk. Pediatr Res 12:998, 1978

17. Harris H, Wirtschafter D, Cassady G: Endotracheal intubation and its relationship to bacterial colonization and systemic infection of newborn infants. Pediatrics 56:816, 1976

18. Miller TC: Emergency treatment of group B streptococcaldisease. Pediatr Clin North Am 24:501, 1977

19. Quirante J, Ceballos R, Cassady G: Group B hemolyticstreptococcal infection in the newborn: Early onset infection.

Am J Dis Child 128:659, 197420. Mims LC, Medawar MS, et al: Predicting neonatal infections

by evaluation of the gastric aspirate: A study in 207 patients.Am J Obstet Gynecol 114:232, 1972

21. Vason V, Lim DM, et al: Origin of gastric aspirate polymorphonuclear leukocytes in infants born after prolonged nipture of membranes. J Pediatr 91:69, 1977

22. Baker CJ: Early-onset group B streptococcal disease. JPediatr93:124,1978

23. Hemming V, McCloskey D, Hill H: Pneumonia in the neonate associated with group B streptococcal septicemia. AmJDis Child 130:1231,1976

Noted by A.G.S.P.

ARTICLES 263

METAPHORS HELP US TO UNDERSTAND THE BODY

“¿�. . . medicine did not make an effective contribution to human welfare until the

middle of the twentieth century. The great leap forward is often attributed to arapid increase in heroic procedures and the discovery of new drugs, but whatdistinguishes the medicine of the past twenty-five years is not that its practi

tioners are equipped with an arsenal of antibiotics and antiseptics, but that theyare furnished with a comprehensive and unprecedented understanding of whatthe healthy body is and how it survives and protects itself. . . . since finding outwhat something is is largely a matter of discovering what it is like, the mostimpressive contribution to the growth of intelligibility has been made by theapplication of suggestive metaphors.. . . It is impossible to imagine how anyone could have made sense of the heart

before we knew what a pump was. Before the invention of automatic gun turrets,there was no model to explain the finesse of voluntary muscular movement.. . . the subjective experience of the body is usually incoherent and perplexing,

and when we want it put right, we refer to people who have learnt to thinkabout it with the help of technical metaphors . . . .“

From Miller J: The Body in Question. New York, Random House, Inc, 1978.

by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from

1980;65;258PediatricsMichael P. Sherman, Boyd W. Goetzman, Charles E. Ahlfors and Richard P. Wennberg

PneumoniaTracheal Aspiration and Its Clinical Correlates in the Diagnosis of Congenital

  

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/65/2/258including high resolution figures, can be found at:

Citations http://pediatrics.aappublications.org/content/65/2/258#related-urls

This article has been cited by 3 HighWire-hosted articles:

Permissions & Licensing

http://pediatrics.aappublications.org/site/misc/Permissions.xhtmlor in its entirety can be found online at: Information about reproducing this article in parts (figures, tables)

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

Online ISSN: 1098-4275.Copyright © 1980 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007.has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it

by Michael Sherman on June 24, 2011pediatrics.aappublications.orgDownloaded from