Prognostic value of the nutmeg lung pattern/lymphangiectasia on fetal magnetic resonance imaging.
Barrera CA, Johnson AM, Rychik J, Biko DM, Degenhardt K, Moldenhauer JS, Victoria T.Pediatr Radiol. 2021 Apr 15. doi: 10.1007/s00247-021-05061-4. Online ahead of print.PMID: 33856503
Take Home Points:
- Nutmeg lung pattern on MRI is an independent risk factor for 30-day mortality with a hazard ratio of 6.1
- This was primarily in patients with CHD (HLHS) with pulmonary vein congestion on fetal echo being the strongest predictor of nutmeg lung
Commentary from Dr. Jared Hershenson (Greater Washington DC), section editor of Pediatric Cardiology Journal Watch:
Pulmonary lymphangiectasia (PL) can be primary or secondary. For fetal cardiologists, most often seen is a secondary form that can develop from congenital heart malformations that generate elevated left atrial pressure and congestion of the pulmonary veins. This is most often seen in fetuses with HLHS and an intact or restrictive atrial septum. Fetal echo findings using antegrade vs. retrograde VTI on the pulmonary venous blood flow can predict those with either poor outcomes or possibly those that should be considered for fetal atrial septostomy/stenting. With the advent of fetal MRI, the finding of nutmeg lung has been considered worrisome for significant obstructive disease. It has additionally allowed for the diagnosis of primary lymphangiectasia that cannot be diagnosed with ultrasound. This pattern may be valuable in counseling as well as aiding in decisions for prenatal therapy.
This retrospective single center study from Children’s Hospital of Philadelphia evaluated 53 patients who underwent fetal MRI due to suspected PL. All scans were reviewed independently by 2 radiologists. Nutmeg lung was defined as a “heterogeneous appearance of the pulmonary parenchyma with T2-hyperintense linear structures radiating peripherally from the hilum to the pleural surface” and primary disease diagnosed in the absence of CHD. Fetal echo was done the same day or earlier than the MRI with specific note of pulmonary vein congestion, reversal of flow, and assessment of the atrial septum. Univariate and multivariate logistic regression models were performed to estimate the association between echo findings and nutmeg lung as well as risk of mortality.
17 fetuses (32%) had nutmeg lung with 12/17 also having CHD, most commonly HLHS. Of those without nutmeg lung, nearly all had CHD; only 1 was diagnosed with primary PL postnatally. 3 fetuses with HLHS and nutmeg lung and with intact atrial septum had atrial septostomy/stenting, only 1 of which was successful. See table 1.
Sixteen fetuses died at a median of 8 days of life, all had CHD (10 with HLHS). Interobserver agreement for nutmeg lung on MRI was excellent (kappa=0.83); 88% had bilateral, 12% right lung only, and 0% left lung only.
Of the fetuses with CHD, 16/47 had pulmonary venous congestion on echo. As expected, a higher proportion of those fetuses had nutmeg lung on MRI (75%), resulting in an odds ratio of 12 on univariate analysis. Of those with HLHS, univariate analysis showed an odds ratio of 19.7 for those with pulmonary venous congestion and 12 in those with intact atrial septum, but multivariate analysis was only significant for pulmonary venous congestion (odds ratio of 13.3). See table 2.
Survival analysis of 52 fetuses (one had intrauterine demise) showed that nutmeg lung was an independent predictor of mortality (HR 6.1), primarily in those with CHD. In fetuses with HLHS, on multivariate analysis nutmeg lung and intact atrial septum were predictors of mortality, but not pulmonary venous congestion. See table 3.
In discussion, the authors mention that for a diagnosis of primary PL, the utility of MRI cannot be understated, as ultrasound can detect the typical pleural effusions but not the diagnosis. While the sample size was small, no fetuses with primary PL died and all had successful neonatal intervention (percutaneous embolization of the abnormal lymphatic channels). In secondary PL, primarily in those with HLHS, nutmeg lung was an independent risk factor of mortality, with Kaplan-Meier curves showing nearly 50% of patients dying or needing transplant by 2 weeks of life. While pulmonary venous congestion was associated with nutmeg lung, it was not predictive of mortality, suggesting that MRI evaluation may be useful for prognostic purposes as well as more urgent discussions for prenatal intervention. Limitations include a small sample size and selection bias. Hopefully more data will become available as fetal MRI is more widely used in the future.
