Systematic literature search using the terms “hypoplastic left heart syndrome” or “HLHS” and “neurodevelopment” and “hybrid” or “norwood” were performed on five electronic biomedical literature databases including Embase, Ovid, Web of science, Scopus, and Cochrane. During analysis, the reference list of each retrieved paper was further scanned for additional studies according to the selection criteria and publications analyzing functional cardiac outcome were added in the analysis. A second literature research was added by using the term (“Hypoplastic left heart syndrome” or “HLHS”) and hybrid and (neurodevelopmental* or “cerebral imaging” or “brain imaging” or “magnetic resonance imaging” or “MRI” or “ultrasound”).

Eligible studies for the review included mono- or multicentric original publications (OPs), with retrospective or prospective study design, analyzing short-to-long-term neurodevelopmental outcome after Hybrid procedure as initial stage I palliation in patients with HLHS as well as the functional cardiac outcome as defined above. Hybrid procedure was defined as combined bilateral pulmonary artery banding with ductal stenting or continuous prostaglandin E1 infusion to preserve ductal patency and balloon atrioseptostomy.

Full OPs were extracted and checked critically for appraisal of evidence. Analyzed factors included name of the first author, year of publication, study design, number of patients, time, type and result of ND assessment, number of patients, type of CHD (HLHS/non-HLHS), type of stage I (Hybrid/Norwood), study assessment, and main findings.

The ND outcome in patients with HLHS undergoing Hybrid procedure has been determined in retrospective (n = 3), prospective (n = 4), monocentric (n = 3), or multicentric (n = 4) studies (Table 1). The number of analyzed patients per study ranged between 16 and 48, but the patient cohorts analyzed in the different studies were overlapping due to subsequent follow-up study with evaluation at an older age (4 years of age), or patient analysis focusing solitary on HLHS/hypoplastic left heart complex (HLHC) patients or solitary treatment by Hybrid and not comparing with Norwood (19, 21). Some studies additionally included infants with a “non-HLHS” diagnosis of CHD defined as HLHC with left heart hypoplasia according to small left heart structures due to dysbalanced atrioventricular septal defect with left ventricular hypoplasia (Table 1). The ND assessment was conducted at different time points starting at 2 months of age until 4 years of age. Most frequently, Bayley II or III scale was used as primary ND outcome parameter.

Overall, the results of ND assessment after Hybrid procedure as stage I were mildly below normal-referenced data of control subjects.

At 2–4 months of age, mild affection of motor development was found using the Test of Infant Motor Performance (TIMP) (37) with a mean TIMP score at 2 months of age of 63.9 ± 18.1 (−1 to −2 SD) and at 4 months of age of 108.3 ± 14.9 (−0.5 to −1 SD) (17).

At the first half year of life (0–6 months), the neuromotor score [modified after (24) with a scaling of 0–18, i.e., 0 normal to 18 severely abnormal] was median (range) 4 (1–9) (19) after the Hybrid procedure.

At 12 months of age, the Bayley Scales of Infant Development II (Bayley II) were lower than the norm of 100 with median psychomotor development index (PDI) and mental development index (MDI) [PDI 57 (49–99), P < 0.001; MDI 91 (65–109), P = 0.002] (19). At the same age (12 months), Bayley III scores were comparable to the population norm for cognitive composite score (CCS) (95.2 ± 8.14) and 1 SD below the norm for language composite score (LCS) (93.6 ± 9.6) and motor composite score (MCS) (82.1 ± 11.9) (18).

At 2–3 years of age, despite mild neurological abnormalities, the median (range) LCS 97 (68–124) was below norm, while CCS with 100 (65–120) and MCS with 97 (55–124) were comparable to normative data (21) (Table 1).

At 4 years of age, cognitive outcome was determined by Wechsler Primary Preschool Intelligence Scale III (WPPSI III) and the Movement-ABC 2 for children after Hybrid were below the norm with median (range) IQ 88 (76–116) (20).

Comparative studies on ND outcome in children comparing Hybrid and Norwood procedure as stage I did not determine differences between cognitive, motor, or language development, neither using Bayley II scale at 1 year of age (19) nor using Bayley III scale at 2–3 years of age (21) or WPSSI III at 4 years of age (Table 1). Comparing the type of CHD (HLHS vs. non-HLHS) treated by Hybrid of ND outcome until preschool age ND outcome was similar (18, 19, 21).

Cerebral imaging findings were analyzed in four studies (Table 2) (17, 25–27). Transcranial Doppler cerebral blood flow (CBF) from the middle cerebral artery (17, 27) and cerebral MRI for assessment of brain volumes (26) showed lower CBF in infants (2–6 months of age) after Hybrid procedure as stage I compared to a control population (27), but showed no correlation with ND outcome at 6 months of age (17). Saiki et al. compared Doppler cerebral blood flow over a longer (chronic) time period between stage I and stage II with reduced oxygenation balance index between lower and upper body in patients after Hybrid procedure. While the total brain volume and total, deep gray, and white matter volumes were reduced in HLHS/C compared to controls at 2–3 years of age, the volume reduction of deep gray and white matter was more reduced after Norwood as stage I procedure compared to Hybrid procedure (26). Of note, intracranial cerebral spinal fluid volume was larger compared to healthy controls before stage III Fontan procedure and correlated with impaired ND outcome (25, 36).

Chan et al. compared in a small retrospective single-center study the weight-for-age and height-for-age (Z-score) somatic growth in three patient cohorts: Hybrid procedure with an early (within 4–6 weeks) switch to Norwood is defined as four-stage procedure (n = 5), Hybrid procedure with later comprehensive stage I/II is defined as a three-stage procedure (n = 8), and Norwood as stage I procedure (n = 49). The decline of weight-for-age (Z-score) between stage I and II was most distinct in the patient group of four-stage followed by three-stage procedure and the Norwood group (28), while in all three groups, there was a catch-up growth until stage III Fontan procedure (Table 2).

Limited pulmonary artery growth was described by smaller pulmonary artery indices after Hybrid procedure compared to Norwood patients before Fontan procedure at 2–3 years of age (29), while in another study, no differences of the pulmonary artery growth parameters such as McGoon ratio at comprehensive stage II and Fontan procedure in patients after Hybrid procedure were shown (30).

Three studies analyzed the myocardial function comparing Hybrid and Norwood procedure using cardiac MRI or echocardiography (31–33). At 2–3 years of age, in cardiac MRI, right ventricle (RV) size was larger after Norwood compared to Hybrid procedure, while in both groups, global RV function was preserved but RV myocardial strain was reduced (32). This was also determined by echocardiography, with no differences until 3 years of age regarding RV size, RV function determined by right ventricular fraction area of change, or calculated global RV radial shortening after Hybrid and Norwood procedure (33) (Table 2).

In the past, ND outcome of patients treated for HLHS has been analyzed by the single ventricle reconstruction trial of the Pediatric Heart Network Investigators (41). This network was started by enrolling a large population of HLHS patients from a number of centers in North America undergoing Norwood procedure with the primary focus comparing the type of aortopulmonary shunt procedure, i.e., the modified Blalock–Taussig shunt and right ventricular-to-pulmonary artery shunt (42). Nevertheless, the trial also assessed the ND outcome for HLHS after Norwood procedure at 14 months (43). In their study, the mean (SD) PDI (74 ± 19) and MDI (89 ± 18) scores were lower than normative means. Compared to the findings after Hybrid procedure at 1 year of age, PDI trended to be lower, while MDI was comparable after Hybrid procedure [PDI 57 (49–99), P < 0.001; MDI 91 (65–109), P = 0.002] (19). At 3 years of age, they found developmental delay in all domains of the Ages and Stages Questionnaire (ASQ) ranging between 17% and 35%, including communication, gross and fine motor, problem solving, and personal social interaction. Furthermore, behavioral attitudes, quality of life, and functional status were also reduced (44). Our review on ND outcome after Hybrid procedure as stage I for HLHS determines an impaired ND outcome in all domains of early childhood development until 3 years of age, comparable to the findings of the Pediatric Heart Network Investigators for the ND outcome after Norwood procedure as stage I (19–23, 26, 30, 31). This includes mild neurological abnormalities with rates between 10% and 35% (Table 1). A clear difference of ND outcome between Norwood and Hybrid procedure has not been determined, so far. Therefore, long-term ND follow-up studies comparing both procedures for stage I until age of adolescence or early adulthood assessing complete neuropsychological profiles with specific notice on cognitive ND outcome measurements such as memory, executive function, and processing speed are needed (45).

Two decades ago, ND outcome was compared between Norwood patients and primary heart transplantation at school age, showing for both groups an overall lower neurocognitive test results (full scale IQ 86 ± 14) without difference between groups stratified by surgical strategy (46). Nowadays, there are some studies determining lower to normal Bayley Scales III; differences between the Hybrid and Norwood patient groups were not evident as well as between the HLHS and non-HLHS patient groups (Table 1). Therefore, it is recommended that children with complex CHD are regularly monitored in a structured ND follow-up program to identify early developmental delay and introduce early interventions, when needed (47).

According to the comparable results of ND outcome at 2–3 years of age after Hybrid and Norwood procedures, this has been also shown for overall mortality rate (including interstage mortality) comparing both treatment strategies. Lloyd et al. analyzed the preoperative condition and interstage mortality in Norwood and Hybrid procedures for HLHS (48). Although the patients after Hybrid procedure had a significantly higher Aristotle score, there was no difference for mortality at any stage. Yerebakan et al. reported single-center results of surgical mortality at stage I, comprehensive stage II, and Fontan procedure of 2.5%, 4.9%, and 0%, respectively, and an interstage mortality of 14.2% leading to a probability of survival of 77.8% at 10 years (30). Baba et al. compared survival after stage II and Fontan procedure after Hybrid and Norwood procedure as stage I, which were equivalent (14).

Differences in myocardial function in patients undergoing Hybrid and Norwood as stage I procedure have been analyzed (Table 2). While Kobayashi et al. focused on acute changes of myocardial function undergoing Hybrid procedure, they showed higher right ventricular myocardial performance index (MPI) in Hybrid patients before intervention, followed by a decrease of MPI after Hybrid procedure within 2 weeks (34). For the chronic changes of myocardial function, Grotenhuis et al. described similar indices of RV size over a longer follow-up for systolic and diastolic function throughout all stages of palliation for Norwood and for Hybrid procedure patients (35). In contrast, right ventricular remodeling seems to be altered when analyzing global radial shortening and deformation indices, as determined by the Toronto group (33). While right ventricular fractional area change remained normal until stage III Fontan procedure independent of Hybrid or Norwood procedure, the global radial shortening increases after stage I in both groups due to greater relative RV circumferential contraction and unchanged longitudinal strain until stage III. Nevertheless, the RV remodeling after Norwood and Hybrid procedure were comparable, and the impact of CPB bypass surgery during stage I was limited (33). Within the largest cohort comparative study between stage II and III, Latus et al. showed comparable preserved global RV pump function determined by cardiac MRI, but better RV strain, strain rate, and intraventricular synchrony after Norwood stage I, both factors not influencing all-cause mortality (32).

Pulmonary artery growth until stage III is more affected after Hybrid compared to Norwood procedure accompanied by a higher number of catheter interventions after Hybrid procedure (Table 2) (14, 18, 30, 32).

Somatic growth (determined as weight-for-age and height-for-age) declines after stage I. This decline continues until stage II and is most prominent for the critical ill neonates, who undergo Hybrid compared to Norwood procedure. Until stage III, Fontan procedure, a moderate increase of body growth follows (28). Nevertheless, after birth restricted body growth has been described for all growth parameters including body weight, body length, and head circumference, determined at birth before stage I, and is followed by a catch-up growth starting after stage II until stage III most prominent for the head circumference which is strongly correlated with the intracranial cerebral spinal fluid volume at 2–3 years of age (25).

On the other hand, brain growth determined for total brain volume and total, deep gray, and white matter volumes is reduced in a mixed population with HLHS and HLHC compared to healthy controls, while deep gray and white matter volumes are more reduced after Norwood stage I compared to Hybrid stage I (26).

Between stage I and stage II, cerebral perfusion determined by transcranial Doppler flow of middle cerebral artery that is either acute within the first week after Hybrid or until stage II is severely impaired after Hybrid (27). Similar findings were found in HLHS patients after Hybrid procedure as stage I with lower cerebral blood flow for 2–6 months compared to controls, but not correlating with ND outcome at 6 months of age (17). The oxygenation balance index between the lower and upper body was lower after Hybrid compared to Norwood procedure as stage I demonstrating impaired cerebral perfusion, which may be due to ongoing retrograde cerebral perfusion after Hybrid procedure as stage I (27). On the other hand, mesenteric perfusion determined by Doppler mesenteric indices improves after Hybrid procedure, which correlates with the potential beneficial effect for the prevention of necrotizing enterocolitis, which is a frequent complication in patients with HLHS by dysbalanced systemic (i.e., cerebral) and pulmonary perfusion (31).

No long-term data comparing ND outcome after Hybrid or Norwood procedure as stage I procedure are available so far, although the technique was developed more than two decades ago. The total number of patients studied remains low for risk factor analysis in a highly multifactorial setting.