Incidence and clinical predictors of Continuous Positive Airway Pressure (CPAP) failure among preterm neonates: a prospective clinical research study protocol.

This prospective clinical research protocol outlines a longitudinal cohort study to be conducted in Dar es Salaam, Tanzania, from March to August 2026, aiming to estimate the incidence and identify clinical predictors of Continuous Positive Airway Pressure (CPAP) failure among preterm neonates using Silverman-Anderson scoring and multivariable linear modeling.

The Gist

The Big Picture: A "Stress Test" for Tiny Lungs

Imagine a newborn baby's lungs are like a brand-new, untested engine in a very small car. Sometimes, especially if the car was built a little early (premature birth), that engine struggles to run smoothly on its own. It needs a little "turbo boost" to get going.

In the medical world, this turbo boost is called CPAP (Continuous Positive Airway Pressure). It's a machine that blows a gentle, constant stream of air into the baby's nose to keep their tiny airways open, acting like a support beam holding up a tent so it doesn't collapse.

The Problem:
Sometimes, even with this "turbo boost," the engine still sputters. The baby's lungs are too weak or too sick to handle the job, and the CPAP machine isn't enough. When this happens, doctors have to switch to a much more invasive method: sticking a tube down the baby's throat to breathe for them. This switch is called "CPAP Failure."

The Goal of This Study:
The researchers in Tanzania want to build a better map of when and why this "engine failure" happens. They aren't just counting how many babies fail; they want to know the speed at which it happens (incidence rate) and the early warning signs (predictors) that tell a doctor, "Hey, this baby is likely to need the tube soon."

---

The Analogy: The "Seven-Day Race"

Think of this study as a seven-day race for newborns in Dar es Salaam, Tanzania.

1. The Start Line: The race begins the moment a baby with breathing trouble is put on the CPAP machine.
2. The Track: The race takes place in three major public hospitals in Dar es Salaam (Amana, Mwananyamala, and Temeke). These are the main "stadiums" for sick babies in the city.
3. The Finish Line: The race ends for a baby in one of two ways:
   • Winning: The baby's lungs get strong enough to breathe on their own, and they stay on CPAP successfully for the first week.
   • Losing (Failure): The baby's breathing gets worse, and they have to be taken off CPAP and put on a ventilator (the tube).
4. The Referees: Every 4 to 6 hours, the medical team checks a "scorecard" called the Silverman-Anderson Score. This is like a referee checking the baby's breathing effort, chest retractions, and grunting. If the score gets too high, it means the baby is struggling.

What They Are Looking For (The Clues)

The researchers are acting like detectives looking for clues that predict who will "lose the race" (fail CPAP). They are collecting data on:

• How early the baby was born (The "engine size").
• How heavy the baby is (The "fuel load").
• How the baby was born (C-section vs. natural).
• How the baby looked right after birth (Apgar scores).
• How much oxygen the baby needed immediately.

They are trying to answer: Is a baby born at 28 weeks more likely to fail than one born at 34 weeks? Does a baby born via C-section have a better chance? Does the amount of oxygen needed in the first hour predict the outcome?

Why This Matters (The "Why")

The paper argues that while we know CPAP is great, we don't have enough fresh, local data from Africa to know exactly how often it fails and why.

• The "Old Maps" Problem: Previous studies from other countries (like Australia or France) or older studies in Tanzania are like using a map from 20 years ago. They might be outdated, too small (not enough babies to be sure), or have design errors.
• The "Referral Bias" Warning: The authors admit a limitation: They are only studying babies at big referral hospitals. This is like studying only the cars that made it to the mechanic's shop. They might miss the babies who were too sick to even get to the hospital, or those who were sent to even bigger hospitals. This might make the failure rate look different than it is in the whole city.

The Plan in Simple Steps

1. Recruit: They will find every baby born in these three hospitals who needs CPAP.
2. Watch: They will watch these babies closely for up to 7 days.
3. Measure: They will write down every detail: the baby's weight, the machine settings, the oxygen levels, and the "stress scores."
4. Analyze: They will use math (statistics) to see which factors are the strongest predictors of failure.

The Bottom Line

This paper is a protocol, which means it is the blueprint or the recipe for a study that hasn't happened yet (it is scheduled for March to August 2026).

The authors are saying: "We are going to run a careful, 7-day observation race in Dar es Salaam to figure out exactly how often CPAP fails and what specific signs tell us a baby is about to fail. We want to replace old, shaky data with new, solid facts so doctors can make better decisions for these tiny patients."

Important Note: The paper explicitly states that this is a plan, not a report of results yet. It does not claim to have found the answers; it only claims to have a solid plan to find them.

Technical Summary

Technical Summary: Incidence and Clinical Predictors of CPAP Failure Among Preterm Neonates

Problem Statement
Acute respiratory failure, frequently manifesting as Respiratory Distress Syndrome (RDS), is a leading cause of neonatal mortality, particularly in sub-Saharan Africa where preterm birth complications are a primary driver of death. While Continuous Positive Airway Pressure (CPAP) is the standard non-invasive intervention for RDS, a significant proportion of neonates fail CPAP therapy, necessitating escalation to invasive mechanical ventilation. This escalation is associated with increased risks of mortality, intraventricular hemorrhage, and bronchopulmonary dysplasia (BPD).

Existing literature regarding CPAP failure rates and predictors suffers from critical limitations. Many studies are statistically underpowered, rely on registry data with imprecise risk adjustments, or are conducted in high-income settings with findings that may not be generalizable to developing nations. Furthermore, previous studies in Africa have been criticized for overt design and analysis errors, often failing to identify independent predictors due to insufficient sample sizes. There is a specific paucity of data estimating the incidence rate (as opposed to prevalence) and identifying clinical predictors of CPAP failure among preterm neonates in representative urban African settings.

Methodology
This paper outlines a protocol for a prospective, longitudinal, observational, analytical cohort study.

• Setting: The study will be conducted across three public regional referral hospitals in Dar es Salaam, Tanzania (Amana, Mwananyamala, and Temeke), which serve as the primary neonatal care facilities for the metropolitan area.
• Population: The target population includes preterm neonates (gestational age ≤37 weeks) delivered within the past six hours who present with signs of RDS and are initiated on CPAP.
• Exclusion Criteria: Neonates with congenital anomalies, those referred to higher-level facilities (e.g., Muhimbili National Hospital), and those requiring only low/high-flow nasal cannula oxygen therapy are excluded.
• Sample Size: A minimum of 196 neonates is required to detect the incidence of CPAP failure with 80% power and a 5% significance level.
• Data Collection: Data will be collected via Case Report Forms (CRFs) over a six-month period (March to August 2026). Assessments include maternal factors (e.g., antenatal corticosteroids, mode of delivery), neonatal demographics (birth weight, gender, Apgar scores), and clinical indicators (Silverman-Anderson scores, FiO2 requirements, CPAP settings).
• Monitoring: Neonates will be monitored every 4–6 hours (or as clinically indicated) for signs of CPAP failure.
• Definition of CPAP Failure: Failure is defined by clinical deterioration requiring escalation of respiratory support, specifically:
  • FiO2 >60% for >1 hour to maintain SpO2 90–95%.
  • ≥3 apnea episodes requiring stimulation within an hour.
  • Any episode requiring bag-mask ventilation.
  • Bradycardia (<100 bpm).
  • Capillary refill >3 seconds.
  • Persistent/worsening respiratory distress (Silverman-Anderson score).
• Statistical Analysis: The primary outcome is the proportion of new CPAP failures per newborn-time of follow-up. Bivariate analyses (Chi-squared, Fisher's, t-tests, or Mann-Whitney U) will identify associations. Variables with p<0.20 will be entered into a multivariable linear model to determine independent predictors. Results will be reported as adjusted absolute risks with 95% confidence intervals.

Key Contributions and Strengths
The study protocol identifies several methodological strengths intended to address gaps in current literature:

1. Incidence vs. Prevalence: Unlike previous studies that primarily reported burden using prevalence estimates, this study aims to provide a precise incidence rate measure of CPAP failure.
2. Granular Stratification: The study intends to characterize incidence and predictors across distinct birth status categories (extreme preterms, preterms, terms, and post-term >42 weeks), a level of differentiation not previously achieved in original findings for this region.
3. Local Context: By focusing on representative public referral hospitals in Tanzania, the study aims to generate data relevant to the resource-limited settings of developing nations, where CPAP failure rates and predictors may differ from high-income registry data.

Results
As this document is a study protocol posted prior to data collection, no empirical results are reported. The paper details the planned analytical framework but does not present findings regarding the actual incidence rate or specific clinical predictors of CPAP failure.

Significance and Claims
The authors claim that this study will provide the first original findings to characterize the incidence rate and clinical predictors of CPAP failure among newborns in Tanzania, stratified by gestational age. The significance of the research lies in its potential to:

• Improve survival chances for newborns with prematurity and RDS by clarifying the nature of standard intervention strategies in developing nations.
• Inform clinical decision-making regarding the timing of escalation from CPAP to mechanical ventilation.
• Contribute to the broader epidemiological understanding of early-life respiratory illnesses as precursors to long-term morbidity and mortality, aligning with demographic transition trends in sub-Saharan Africa.

The study acknowledges limitations, including potential referral bias due to the exclusive use of referral hospitals and the risk of high attrition rates due to early neonatal deaths. Ethical clearance has been obtained from Kairuki University, and informed consent will be secured from mothers of all recruited neonates.