Diaphragm dysfunction (DD) following congenital heart surgery has been described as an iatrogenic phrenic nerve injury and a critical factor in failure of postoperative functional recovery. The classic diagnostic triad consists of difficulty weaning from mechanical ventilation, increased work of breathing, and persistent atelectasis. However, clinical practice has been characterized by a remarkable lack of standardization: although the pathophysiology is well understood, the timing of diagnosis and treatment has historically ranged from expectant management to early plication, often driven more by institutional experience than by robust evidence.
Until now, anatomical and procedural risk factors had been identified. However, a gap remained in decision-making, particularly in the most complex patients, in whom the distinction between transient phrenic nerve palsy and clinically significant dysfunction requiring surgical intervention remained unclear. The recent publication by Ruppe et al. in The Journal of Thoracic and Cardiovascular Surgery (2026) provides an updated perspective on this paradigm. By analyzing a large multicenter cohort from the Pediatric Cardiac Critical Care Consortium (PC4) registry, this study offers a new perspective: it not only confirms the relatively low incidence (1.5%), but also reframes the triggers for intervention, shifting from acting in response to respiratory failure toward an approach based on patient profile and procedural complexity.
The analysis by Ruppe et al. of the PC4 cohort (2014-2022) was based on 77383 patients and places the incidence of this complication at 1.5%. Although this figure may appear low at first glance, the situation becomes considerably more concerning when the highest-risk subgroups are considered.
The study provides clear risk stratification:
- Vulnerability profile: Neonatal age (incidence 2.8%) and surgical complexity (STAT categories 4-5) are among the strongest independent predictors. In our practice, this reinforces the notion that patients undergoing functionally univentricular surgery, particularly at the Norwood or Glenn stages, not only have a higher risk of developing DD, but also represent a group in whom its impact on morbidity and mortality may be greatest.
- Diagnosis and treatment: Perhaps the most striking finding is the variability in diaphragm plication (DP) rates across centers, ranging from 0% to 100%. This heterogeneity suggests that, in the absence of universally accepted clinical guidelines, the decision to perform plication remains highly center-dependent.
- Timing of intervention: The median time to diagnosis was 7 days and the median time to plication was 14 days. However, the observation that centers diagnosing DD before extubation also intervene surgically much earlier suggests that early clinical suspicion, particularly in high-risk neonates, may be key to reducing prolonged cardiac intensive care unit (CICU) stays.
Current evidence confirms that DD is a “marker of vulnerability” in the postoperative setting. Its presence is associated with prolonged hospital stays and an incidence of major complications approximately twice that observed in patients without dysfunction (42.2% vs 21.0%). The question raised by this analysis is how long we are willing to “wait” before the clinical cost in terms of days of mechanical ventilation and associated complications becomes irreversible. DD remains an underdiagnosed complication with a significant impact on morbidity and mortality in these patients.
COMMENTARY:
From a surgeon’s perspective, this study invites us to reconsider the threshold for intervention. We often find ourselves trapped in an expectant approach: when DD is suspected, we wait for days to see whether the patient can be extubated, prioritizing avoidance of an additional surgical procedure over the potential cost of prolonged mechanical ventilation.
Diaphragm dysfunction can be understood as a marker of postoperative vulnerability. Its clinical impact depends not only on the degree of phrenic nerve injury, but also on the respiratory and hemodynamic reserve of the individual patient. Unilateral paresis may be well tolerated by an older child with biventricular circulation yet become a decisive barrier to extubation in a neonate with univentricular physiology.
The data from Ruppe et al. provide an important lesson: failure to extubate should not simply be regarded as “bad luck,” but rather as a clinical profile that warrants early diagnostic suspicion. In daily practice, the key should be to identify the “point of no return.” The reality is that patients undergoing univentricular surgery or highly complex procedures have very limited respiratory reserve. Delaying either the diagnosis of DD or the decision to perform plication not only increases the duration of mechanical ventilation but also adds to the burden of associated complications, such as pneumonia or the need for tracheostomy, ultimately making an already complex postoperative course even more difficult.
The study does not demonstrate that early plication improves outcomes, nor does it establish a specific postoperative day on which plication should be performed. However, the variability between centers reveals that we also lack shared criteria for deciding how long to wait. This is probably the most uncomfortable finding of the study: despite dealing with a well-recognized complication, management still depends to a large extent on where the patient is treated.
Personally, I believe we should stop viewing diaphragm plication solely as a rescue procedure and begin considering it as a tool for early optimization. If diaphragmatic ultrasound assessment can be systematically incorporated after surgery, particularly in the most vulnerable patients, we may be able to intervene before the patient becomes exhausted. Ultimately, the decision to intervene should not represent a failure of conservative management, but rather a planned strategy aimed at shortening CICU stay and returning the patient to baseline.
In conclusion, diaphragm dysfunction following congenital heart surgery remains an underdiagnosed complication with a significant impact on morbidity and mortality. The evidence provided by this study shows that, although its overall incidence is 1.5%, the risk is concentrated in well-defined subgroups: neonates, patients undergoing highly complex procedures (STAT 4-5), and, particularly, those with functionally univentricular anatomy. The substantial variability in plication rates across centers suggests that we remain far from a universal consensus. Integrating early diagnostic suspicion and systematic diaphragmatic ultrasound assessment may allow us to identify patients who could benefit from earlier intervention.
Ruppe et al. do not tell us the exact day on which plication should be performed, but they do show that waiting without a common strategy leads to extreme differences between centers. The next step is not to plicate every patient earlier, but to recognize dysfunction sooner, define how long we are prepared to wait for each patient profile, and establish explicit criteria for abandoning observation. Because in the vulnerable congenital heart patient, waiting is also an intervention.
REFERENCE:
Ruppe M, DeWitt A, Schumacher KR, Jacobs JP, Algaze CA, Smith A, et al. Diaphragm dysfunction following congenital heart surgery: Epidemiology and outcomes. J Thorac Cardiovasc Surg. 2026 Jul;172(1):200-208.e3. doi: 10.1016/j.jtcvs.2026.02.024. Epub 2026 Feb 28. PMID: 41765325.
