Abstract STUDY QUESTION Are ICSI outcomes impaired in cases of severe asthenozoospermia with multiple morphological abnormalities of the flagellum (MMAF phenotype)? SUMMARY ANSWER Despite occasional technical difficulties, ICSI outcomes for couples with MMAF do not differ from those of other couples requiring ICSI, irrespective of the genetic defect. WHAT IS KNOWN ALREADY Severe asthenozoospermia, especially when associated with the MMAF phenotype, results in male infertility. Recent findings have confirmed that a genetic aetiology is frequently responsible for this phenotype. In such situations, pregnancies can be achieved using ICSI. However, few studies to date have provided detailed analyses regarding the flagellar ultrastructural defects underlying this phenotype, its genetic aetiologies, and the results of ICSI in such cases of male infertility. STUDY DESIGN, SIZE, DURATION We performed a retrospective study of 25 infertile men exhibiting severe asthenozoospermia associated with the MMAF phenotype identified through standard semen analysis. They were recruited at an academic centre for assisted reproduction in Paris (France) between 2009 and 2017. Transmission electron microscopy (TEM) and whole exome sequencing (WES) were performed in order to determine the sperm ultrastructural phenotype and the causal mutations, respectively. Finally 20 couples with MMAF were treated by assisted reproductive technologies based on ICSI. PARTICIPANTS/MATERIALS, SETTING, METHODS Patients with MMAF were recruited based on reduced sperm progressive motility and increased frequencies of absent, short, coiled or irregular flagella compared with those in sperm from fertile control men. A quantitative analysis of the several ultrastructural defects was performed for the MMAF patients and for fertile men. The ICSI results obtained for 20 couples with MMAF were compared to those of 378 men with oligoasthenoteratozoospermia but no MMAF as an ICSI control group. MAIN RESULTS AND THE ROLE OF CHANCE TEM analysis and categorisation of the flagellar anomalies found in these patients provided important information regarding the structural defects underlying asthenozoospermia and sperm tail abnormalities. In particular, the absence of the central pair of axonemal microtubules was the predominant anomaly observed more frequently than in control sperm (P < 0.01). Exome sequencing, performed for 24 of the 25 patients, identified homozygous or compound heterozygous pathogenic mutations in CFAP43, CFAP44, CFAP69, DNAH1, DNAH8, AK7, TTC29 and MAATS1 in 13 patients (54.2%) (11 affecting MMAF genes and 2 affecting primary ciliary dyskinesia (PCD)-associated genes). A total of 40 ICSI cycles were undertaken for 20 MMAF couples, including 13 cycles (for 5 couples) where a hypo-osmotic swelling (HOS) test was required due to absolute asthenozoospermia. The fertilisation rate was not statistically different between the MMAF (65.7%) and the non-MMAF (66.0%) couples and it did not differ according to the genotype or the flagellar phenotype of the subjects or use of the HOS test. The clinical pregnancy rate per embryo transfer did not differ significantly between the MMAF (23.3%) and the non-MMAF (37.1%) groups. To date, 7 of the 20 MMAF couples have achieved a live birth from the ICSI attempts, with 11 babies born without any birth defects. LIMITATIONS, REASONS FOR CAUTION The ICSI procedure outcomes were assessed retrospectively on a small number of affected subjects and should be confirmed on a larger cohort. Moreover, TEM analysis could not be performed for all patients due to low sperm concentrations, and WES results are not yet available for all of the included men. WIDER IMPLICATIONS OF THE FINDINGS An early and extensive phenotypic and genetic investigation should be considered for all men requiring ICSI for severe asthenozoospermia. Although our study did not reveal any adverse ICSI outcomes associated with MMAF, we cannot rule out that some rare genetic causes could result in low fertilisation or pregnancy rates. STUDY FUNDING/COMPETING INTEREST(S) No external funding was used for this study and there are no competing interests. TRIAL REGISTRATION NUMBER N/A.