Article Text

Download PDFPDF

Long term follow-up of very low birthweight infants from a neonatal volume versus pressure mechanical ventilation trial
  1. J Singh1,
  2. S K Sinha1,
  3. E Alsop1,
  4. S Gupta1,
  5. A Mishra2,
  6. S M Donn3
  1. 1
    Department of Neonatology, James Cook University Hospital, Middlesbrough, UK
  2. 2
    Department of Neonatology, Hope Hospital, Manchester, UK
  3. 3
    Division of Neonatal Perinatal Medicine, C.S. Mott Children’s Hospital University of Michigan Health System, Ann Arbor, Michigan, USA
  1. Correspondence to Professor Sunil K Sinha, Department of Paediatrics and Neonatal Medicine, University of Durham and James Cook University Hospital, Marton Road, Middlesbrough TS4 3BW, UK; sunil.sinha{at}


Background: A previous randomised trial showed volume controlled ventilation (VCV) was efficacious in ventilating very preterm and extremely low birthweight babies.

Objective: To compare long term survival, pulmonary morbidities and gross neurodevelopmental outcomes of babies randomised to either VCV or pressure limited ventilation (PLV) for treatment of respiratory distress syndrome.

Design/Methods: Masked evaluation of health status, including frequency of respiratory illness, use of medications, hospital admissions, and gross neurodevelopmental status were obtained using a structured parental questionnaire and verification from medical records.

Results: 94 of 109 children (86%) survived to discharge. Three died after discharge (2 VCV, 1 PLV). Modality of ventilation did not affect overall mortality; seven VCV children died (12%) versus 11 PLV (21%) (OR 0.5 (95% CI 0.1 to 1.4), p = 0.13). Respiratory abnormalities were present in 32 (37%), and 26 (30%) required hospital readmission. There was no significant difference in readmission rates between the two groups: VC 13/45 (29%) and PLV 19/40 (47%) (OR 0.4 (0.1 to 1.1), p = 0.07). Modality of ventilation did not affect frequency of respiratory illness: VC 12 (27%) and PLV 14 (35%) (OR 0.46 (0.1 to 1.1), p = 0.09). However, significantly fewer VCV children (13%, n = 6) compared to PLV children (32%, n = 13) required treatment with inhaled steroids/bronchodilators (OR 0.3 (0.1 to 0.9), p = 0.04). Nine children had severe neurodevelopmental disability (cerebral palsy, blindness, deafness) (9.8%; 3 VCV, 6 PLV 6) (OR 0.4 (0.09 to 1.7)).

Conclusions: The efficacy of VCV in very preterm and low birth babies appears to be maintained on longer term evaluation.

Statistics from

Mechanical ventilation of the newborn has been traditionally accomplished using time cycled pressure limited ventilation (PLV). This modality of ventilation, however, is associated with inconsistency in tidal volume delivery which may be undesirable in preterm babies. In addition, over-expansion (volutrauma) and/or under-expansion/collapse (atelectotrauma) can contribute to ventilator induced lung injury.1 This has led to increasing interest in volume controlled ventilation (VCV)2 in which the primary gas delivery target is tidal volume and inspiratory pressure is automatically adjusted from breath to breath to deliver the desired tidal volume. Use of other volume targeted modalities, which are essentially pressure regulated but aim to deliver a desired tidal volume set by the operator, such as volume guarantee (VG) and pressure regulated volume controlled (PRVC) ventilation, have also been utilised.3 4

What this study adds

This study provides longer term follow-up data on volume controlled ventilation and strengthens the need for a large, adequately powered randomised trial.

We recently reported the results of a novel, randomised controlled trial of VCV compared to PLV in a population of preterm babies with respiratory distress syndrome (RDS), who were born between 24 and 31 weeks’ gestation weighing 600–1500 g. In this study, VCV was found to be associated with faster weaning, reduction in duration of ventilation and improved survival, especially in the subgroup of babies weighing between 600 and 1000 g.5

The aim of the present study was to assess the outcomes of survival and respiratory and gross neurodevelopmental status at around 2 years of age as part of routine clinical follow-up. These parameters had been defined, a priori, as secondary outcomes when the original study was planned. The respiratory outcomes of interest were the occurrence and frequency of cough and wheeze, the use of inhaled medications for respiratory dysfunction, and hospital admissions for respiratory illnesses. The neurodevelopmental outcome of interest was the presence of severe disability, defined as cerebral palsy severe enough to hamper gross motor activity appropriate for the child’s chronological age, deafness needing hearing aids, or blindness (registered blind or partially sighted).


Study population

The original study protocol has been described in detail previously.5 In summary, the study recruited 109 babies of 24–31 weeks’ completed gestation, weighing 600–1500 g, who had RDS requiring mechanical ventilation and surfactant treatment. All babies were randomised within 6 h of initiation of mechanical ventilation to receive either PLV or VCV. The primary outcome of the study was the speed of weaning assessed by predefined criteria (mean airway pressure less than 8 cm H2O and alveolar-arterial oxygen gradient less than 13 kPa or 100 mm Hg). Other outcome measures included duration of mechanical ventilation, survival to discharge, and complications generally associated with prematurity. All surviving children were to be prospectively followed by their primary paediatrician as part of their routine care. Evaluation of health status and gross neurodevelopmental status were obtained from the follow-up medical assessments by paediatricians and through parental interviews using a structured questionnaire. Data were collected using a modified version of a questionnaire previously validated in a randomised trial comparing two surfactants.6 Data recorded included the occurrence and frequency of cough and wheeze, use of respiratory medications, home oxygen use, and hospital admissions (for both respiratory and other reasons). The person administering the questionnaire was masked to the modality of ventilation to which the child had been assigned.

Approval for the study was granted by the James Cook University Hospital Research Ethics Committee.

Statistical analyses

We compared infant and maternal variables to the original demographics to confirm that deaths or loss to follow-up had not affected the balance of the two groups. Analysis was on an intention-to-treat basis. Odds ratios (OR) with 95% confidence intervals (95% CI) were calculated to estimate the relative effect of VCV compared to PLV for all categorical outcomes using logistic regression. Adverse neurological outcomes were uncommon and were dichotomised to severe disability or any other outcome. On the other hand, respiratory morbidity was common and each variable was dichotomised to any adverse outcome or normal outcome. The statistical software used was SPSS for Windows.


Of the 94 infants surviving to discharge, three subsequently died. No respiratory or neurological outcome data were available for six survivors, leaving 85 subjects (93%) with complete datasets (fig 1). One child died at 6 weeks’ corrected age from group B streptococcus infection, the second death occurred at 6 months’ corrected age from chronic lung disease, and the third death resulted from non-accidental injury. Overall, there were seven deaths in the VCV group (12%) compared to 11 (21%) in the PLV group. Modality of ventilation per se did not effect mortality (OR 0.5 (95% CI 0.1 to 1.4), p = 0.13).

Figure 1

Recruitment and follow-up of children.

Infant and maternal characteristics between the two groups were quite similar for the major determinants of outcome: birthweight, gestational age, gender, or major abnormality on cranial ultrasound scan (table 1). Oxygen dependency at 36 weeks’ postmenstrual age, and the need for supplemental oxygen at discharge did not differ significantly between the two groups. The median age at follow-up was 22 months (interquartile range (IQR) 16–26) for the whole study group. The median time of assessment in the VCV group was slightly earlier (median 18 months, IQR 16–24) compared to the PLV group (median 24 months, IQR 18–26) The frequency of reported respiratory symptoms was high: 32/85 (37%) of parents reported that their child had cough or wheeze. Overall, 19/85 (22%) had received inhaled medication for cough and wheeze. Hospital admissions for respiratory problems were common (26/85, 30%). When the two study groups were compared, for those children assigned to VCV, the odds ratio for respiratory symptoms was 0.4 (0.1 to 1.1), p = 0.09 (table 2) and for use of inhaled medications it was 0.32 (0.1 to 0.9), p = 0.04. There was no significant difference in the rate of hospital admissions secondary to respiratory or other causes.

Table 1

Characteristics of study population

Table 2

Respiratory outcomes and subsequent hospitalisations

Three children in the VCV group had severe disability compared to six in the PLV group (OR 0.4 (0.09 to 1.7)). Their patterns of disability are shown in table 3. The combined outcome of death or disability was not significantly different in the two groups: 10/57 (17.5%) in the VCV group compared to 17/52 (32.6%) in the PLV group (p = 0.067). Six children in the PLV group had seizures compared to three in the VCV group (p = 0.29).

Table 3

Patterns of severe disability of children in the two groups


This study is unique in following the longer term outcomes of infants enrolled in a neonatal randomised clinical trial of volume targeted versus pressure targeted ventilation. In the original trial, we had found that VCV was an effective method of ventilating very preterm and extremely low birthweight babies. The follow-up study provides further support for this finding.

Current hospital-based information was used to identify disabilities in survivors. Many of the children in this study were outborn or transferred in utero to receive intensive care. Follow-up of these children after discharge was thus provided by local paediatricians and not by the study centre team. The median time of assessment was slightly different in the two groups. The children in the PLV group were assessed at a later time (median 24 months) than children in the VCV group (median 18 months). Although this was not done deliberately, it might be a confounding factor for frequency of respiratory outcomes and hospital admissions. Overall, there was an excellent ascertainment rate (90%) in surviving children, minimising any response bias. The distribution of perinatal variables such as chronic lung disease at 36 weeks and the need for oxygen at discharge from neonatal intensive care for children lost to follow-up was similar for both groups.

Overall respiratory morbidity was high as has been previously reported for this population.7 Hospital admissions for respiratory problems as assessed clinically, were common but no significant differences were found with regard to children with cough or wheeze. There was also no significant difference in the distribution of children admitted to a hospital for respiratory symptoms. However, the proportion of children using inhaled medications was significantly less in the VCV group. This may be a clinical correlate, as fewer children in the original trial who had received VCV were noted to have chronic lung disease at discharge compared to those who had received PLV. This should be interpreted with caution, however, as the original study was not powered to look at long term outcomes. We realised at the outset that a randomised controlled trial of this size would not definitively prove whether or not one method was superior to the other. We felt that it was imperative to demonstrate the safety and efficacy of this new technique before larger trials were initiated.

The follow-up neurological assessment did not include formal developmental testing. Instead, cerebral palsy severe enough to hamper gross motor activity appropriate for the child’s chronological age, blindness, and profound hearing loss needing hearing aids were chosen as severe disabilities. This definition has been used in previous studies.8 There was no significant difference in the incidence of severe disability in the two groups. The combined incidence of death or disability was also not significantly different in the two groups. We did not attempt to assess the cognitive development of studied children, as it were likely to be non-predictive at this age.

Based on the findings of our studies, the early use of VCV in extremely preterm infants appears to be effective and may have certain advantages. The stage is now set for a larger, adequately powered multicentre trial to compare volume targeted ventilation to pressure limited ventilation to examine clinically relevant long term outcomes.9



  • Competing interests None.

  • Ethics approval James Cook University Hospital Research Ethics Committee.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.