Post-publication Peer Review (P3R) is an online forum for ongoingreview peer review. To submit a P3R please go to the article you wish to respond to and click on the link that reads "P3Rs: Submit a Response." Submission of P3Rs are open to all health care professionals and experts in related fields.

Post-publication Peer Reviews to:

ARTICLE:
Dario Prais, Tommy Schonfeld, and Jacob Amir
Admission to the Intensive Care Unit for Respiratory Syncytial Virus Bronchiolitis: A National Survey Before Palivizumab Use
Pediatrics 2003; 112: 548-552 [Abstract] [Full text] [PDF]
*P3Rs: Submit a response to this article

P3Rs published:

[Read P3R] Outcomes of severity in The Impact Study: Statistics by admission
Carlos Ochoa Sangrador, Javier Gonzalez de Dios   (27 December 2003)

Outcomes of severity in The Impact Study: Statistics by admission 27 December 2003
Top
Carlos Ochoa Sangrador,
MD
Department of Pediatrics. Virgen de la Concha Hospital. Zamora (Spain).,
Javier Gonzalez de Dios

Send letter to journal:
Re: Outcomes of severity in The Impact Study: Statistics by admission

cochoas{at}meditex.es Carlos Ochoa Sangrador, et al.

Outcomes of severity in The Impact Study: statistics by admission

Outcomes of severity in The Impact Study: Statistics by admission

To the Editor:

The paper of Prais et al 1 prove that the great majority of infants admitted to the PICU for bronchiolitis were not candidates for RSV prophylaxis, by means of a prospective multihospitalary survey before the introduction of palivizumab in Israel.  They concluded that new risk-stratified guidelines for RSV prophylaxis are needed, less wide than the actually recommendations of the American Academy of Pediatrics.

In the discussion of the article, the authors say that the results of IMpact study seem to contradict their data of ICU admission (3% of the placebo-treated children in contrast to 1,3% of the palivizumab group). The aim of this letter is recalculate the outcomes of severity in the Impact study (need and days of ICU admissions, need and days for mechanical ventilation, mortality) by means of  statistics by admission (and not for 100 children), as a result of critical appraisal of original clinical trial 2 according to the principles of Evidence-Based Medicine Working Group.3,4

From a critical assessment of the IMpact Study, it is worth questioning if the “primary end points” chosen by the authors (we suppose it was chosen previously) are the most important for the clinician: may it be accepted that severe bronchiolitis are admitted to hospital and therefore a reduction of hospitalization implies a protection of the serious forms? This aspect is important since the outcomes should be coherent with the existence of more moderate forms in the treated regarding to those of the placebo group (which is questionable after the new analysis we propose).

As Prais et al suggest,1 the main impact of the treatment should be focused on the reduction of the severe forms of bronchiolitis. However, the outcomes related to greater severity such as necessity of intensive care, oxygenotherapy, mechanical ventilation and mortality, were considered “secondary end points” in the IMpact Study.

The outcomes of these secondary variables were expressed as cumulative incidence and number of events (days of hospitalization or mechanical ventilation or intensive care, etc) per 100 patients. This analysis may be confusing since all the indicators curiously are apparently favourable for the group treated with palivizumab, when all of them agree that the treated patients hospitalized, even though they are fewer, they are worse. If we calculate the number of days of intensive care, with mechanical ventilation or with supplemental oxygen or with lower respiratory illness/infection score ³3; all the calculations are against the treatment (table 1). In conclusion, the treated seem to be less hospitalized but they are worse, although from a first reading of the article as to secondary variables, it seems quite the reverse.

Paradoxically, although there is no greater mortality, two patients who were treated died during hospitalization (one died during surgery unrelated to RSV infection). The authors argue in a letter5 to the director that there were a reduced number of cases in the treated group with a serious prognosis due to important comorbidity and they got worse (according to them, not due to the treatment; three cases took 60% of intensive care days and 65% of days with mechanical ventilation) what influenced the statistics. However, these circumstances are part of the laws of the game and all should be taken into account in a far-reaching clinical trial as the IMpact Study.

 

Sincerely

C. Ochoa Sangrador, J González de Dios*

Department of Pediatrics. Virgen de la Concha Hospital. Zamora (Spain). *Departament  of  Pediatrics.  San Juan Universitary  Hospital. Miguel Hernández University. Alicante (Spain).

 

References

1.- Prais D, Schonfeld T, Amir J, for the Israeli RSV Monitoring Group. Admission to the intensive care unit for respiratory syncytial virus bronchiolitis: A national survey before palivizumab use. Pediatrics 2003; 112: 548-552.

2.- The Impact-RSV  Study Group. Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants. IMpact-RSV Study Group. Pediatrics 1998; 102:531-7.

3.- Guyatt GH, Sackett FL, Cook DJ. User´s guides to the medical literature. II. How to use an article about therapy or prevention: A. Are the results of the study valid? JAMA 1993; 270: 2598-2601.

4.- Guyatt GH, Sackett FL, Cook DJ. User´s guides to the medical literature. II. How to use an article about therapy or prevention: B. What are the results and will they help me in caring for my patients? JAMA 1994; 271: 59-63.

5.- Connor EM, Carlin D, Top FH, Weishman LE. Who shall not receive palivizumab?. Pediatrics 2000; 106: 866-867.

 

Table I:  Analysis per hospitalized patient. Recalculated from the original data (per 100 children) of the IMpact Study2.    

 

 

 

Analysis per 100 patients a

 

Analysis per hospitalized patient b

Nº patients / hospitalized

 

Placebo

(500)

Palivizumab

(1002)

 

Placebo

(53)

Palivizumab

(48)

Days of hospitalization

 

62.6

36.4

 

5.9

7.6

Increased oxygen (days)

 

50.6

30.3

 

4.8

6.3

LRI score ³3 (days)

 

47.4

29.6

 

4.5

6.2

ICU admissions (Nº / %)   

 

15(3.0%)

13(1.3%)

 

15(28%)

13(27%)

Days of ICU 

 

12.7

13.3

 

4.2c

10.5c

Mechanical ventilation (Nº / %)

 

1(0.2%)

7(0.7%)

 

1(7%)c

7(54%)c

Days of MV      

 

1.7

8.4

 

8.5d

12.02d

Mortality (Nº / %)

 

5(1%)

4(0.4%)

 

0(0%)

2(4.1%)

a  data published in the article of IMpact Study2    

b  data calculated from the data of the article in this way:  to multiply the average number of days of each complication (per 100 children) of each group by the number of hundreds of children of each group and to divide the result (total days) by the number of children of each group with the complication.  For example:  the average stay of 5.9 days in the placebo group is obtained by multiplying the average stay per 100 children in the placebo group (62.6 days) by 5 (since there are 500 children with placebo;  500/100=5), and to divide it by the number of hospitalized children of the placebo group (53).

c considering the patients in intensive care unit    

d considering the patients with mechanical ventilation.

 

LRI= Lower Respiratory Tract Illness/Infection; ICU= Intensive Care Unit; MV= mechanical ventilation.