ABSTRACT
Aim
To assess postmortem investigations performed in sudden unexpected death in infancy (SUDI) cases in daily practice in France.
Methods
Children deceased from SUDI and registered in the French SUDI registry from 2015 to 2024 were included retrospectively. The postmortem investigations performed in SUDI cases were described, their temporal trends since 2015 and their compliance with the French National Health Authority guidelines were assessed. The minimum set of non‐invasive postmortem investigations performed in cases required to detect unnatural death, including fundus examinations, imaging and toxicological tests, was studied.
Results
We examined 1 934 SUDI cases. Laboratory tests were the most frequently performed procedures (93.0%), and fundus examinations were the least frequent (21.6%). There was a significant increase in fundus examinations (9.3% to 31.6%), laboratory tests (87.6% to 97.3%), bacteriological tests (91.4% to 97.2%), toxicological tests (79.0% to 91.0%), and whole‐body imaging (75.2% to 87.3%) from 2015 to 2024. Only 2.1% of cases were investigated strictly according to the French guidelines. We found that 13.2% of cases underwent minimal non‐invasive postmortem investigation for detecting unnatural death.
Conclusion
Although certain investigations are performed more frequently, efforts to improve local protocols and updated guidelines are needed to ensure thorough postmortem investigations, resulting in an improved diagnostic approach.
Keywords: postmortem investigation, sudden infant death syndrome, sudden unexpected death in infancy, sudden unexpected death in infancy registry
Summary
The study aims to determine how the 2007 French National Health Authority guidelines on the management of sudden unexpected death in infancy (SUDI) were effectively implemented.
Analysis of 1 934 SUDI cases from the French SUDI registry from 2015 to 2024 showed that, despite significant improvements, only 2.1% were fully investigated in accordance with the French guidelines.
Efforts to improve local protocols and update guidelines are needed to ensure thorough postmortem investigations.
Abbreviations
- CT
computed tomography
- SIDS
sudden infant death syndrome
- SUDI
sudden unexpected death in infancy
1. Introduction
Sudden unexpected death in infancy (SUDI) is defined as the sudden and unexpected death of an infant under one year of age, the cause of which is not immediately obvious. After exhaustive postmortem investigations, SUDI can be attributed to one or more causes or be classified as sudden infant death syndrome (SIDS) when the death remains unexplained [1, 2]. When explained, aetiologies of SUDI are mainly natural diseases such as acute infections, including pulmonary, central nervous system, or gastrointestinal infections; undiagnosed congenital diseases, including heart defects; metabolic congenital or acquired disorders; accidental asphyxiation; and, to a lesser extent, injuries secondary to fatal child abuse [2, 3]. Studies published from 2008 to 2025 reported that between 33% and 91% of these deaths remain unexplained [3, 4, 5]. Thus, identifying the cause of death remains a crucial issue to support and accompany parents in their bereavement and implement prevention strategies [6].
Although there is currently no internationally validated consensus, many countries are attempting to standardise the protocol for investigating SUDI cases [6, 7, 8, 9]. In 2007, the French National Health Authority issued recommendations to standardise postmortem investigations in SUDI cases throughout the French territory, aiming at maximising the likelihood of identifying a possible cause of death in all cases [10]. The French guidelines state that all infants under two years of age and their families should systematically be referred to a SUDI referral centre where all postmortem investigations, including medical history taking, death scene examination, clinical examination, microbiological and toxicological tests, imaging and proposal medical autopsy, can be performed. Despite these recommendations, a 2009 French survey by the National Public Health Agency revealed significant heterogeneity in practices among referral centres [11]. In 2013, another study on the SUDI investigations proposed by referral centres showed that none of them were able to apply the French guidelines in their entirety [12].
In parallel, a French registry was set up in 2015 to improve the management of SUDI cases, compile comprehensive data and monitor their prevalence in France. It also aims to improve medical knowledge through the collection of standardised and accurate epidemiological data [13, 14]. This study aimed to assess the postmortem investigations performed in SUDI cases in daily practice in France based on the French SUDI registry.
2. Methods
2.1. Study Design and Population
A multicentre retrospective study was conducted based on data from the French SUDI registry. This registry compiles information on all infants under the age of two years who died unexpectedly, in accordance with the international definition of SUDI [1]. All children who died from SUDI and were registered in the French SUDI registry between 15 May 2015 and 30 June 2024 were included. SUDI cases that did not undergo any postmortem investigation and/or whose parents refused inclusion in the French SUDI registry were excluded. This study was reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Table S1) [15].
2.2. Data on Postmortem Investigations
Data relating to postmortem investigations detailed in the protocol proposed in the French guidelines were investigated [10]. The collected data included demographics, site of admission, weight, height, head circumference, body temperature, fundus examination, laboratory tests, bacteriological tests, virological tests, toxicological tests, imaging, and autopsy (Table S2).
The reasons reported by the paediatricians for not performing the postmortem investigations were collected. These were classified into eight categories: examination outside the scope of the local protocol, failure to perform the examination, examination not requested, parental refusal, police inquiry, refusal or absence of specialist physicians, administrative issues, and no information.
2.3. The French Guidelines for Investigating SUDI Cases
Complete and exhaustive compliance with the French guidelines was considered when the weight measurements, size measurements, head circumference measurements, body temperature measurements, fundus examination, bacteriological tests, virological tests, toxicological tests, chest, skeletal X‐rays, brain imaging, whole‐body imaging, and an autopsy were performed. If any of these measurements or examinations were not performed or were missing, the case was considered non‐compliant with the French guidelines.
2.4. Classification of Causes of Death
The causes of death were established by the referral paediatrician and divided into three categories following the classification of the 3rd International Congress on Sudden Infant and Child Death [16]. Death by unexplained cause (SIDS group) was defined as insufficient evidence for defining the cause of death despite complete investigations. The explained group was defined by the presence of suffocation/asphyxia, infectious, cardiac, metabolic, traumatic, toxic, genetic, digestive, other, or multifactorial causes. The undetermined group was defined by incomplete investigations or conflicting evidence for competing causes.
2.5. Statistical Analysis
First, the characteristics of SUDI cases and the postmortem investigations performed throughout the study period were described both in the whole cohort and according to the cause of death. Count and percentage for the qualitative data or median and interquartile range (IQR) for the quantitative data were used. Second, the temporal trends observed for postmortem investigations from 2015 to 2024 were assessed using the Cochran‐Armitage test. Third, compliance with the French guidelines was investigated. Lastly, the minimum set of non‐invasive postmortem investigations performed in SUDI cases required to detect non‐natural death was studied. These investigations included fundus examination, skeletal X‐rays, brain or whole‐body imaging and toxicological tests. For all analyses, R V.4.3.3 software (R Foundation for Statistical Computing, Vienna, Austria) was used.
3. Results
3.1. Population
Between 15 May 2015 and 30 June 2024, there were 2020 SUDI cases registered in the French SUDI registry. After excluding parental refusal (n = 81) and cases that did not undergo any postmortem examinations (n = 5), there were 1 934 (95.7%) cases included in this study.
The median age at the time of death was 3 months (IQR: 1–6). Globally, 1098 infants (57.2%) were boys, and 334 (20.3%) were born at full term (Table 1). SUDI cases were mainly referred to a SUDI referral centre (82.4%), followed by a forensic institute (10.5%) or a hospital that was not a referral centre (7.1%).
TABLE 1.
SUDI cases included in the whole cohort and according to the site of admission.
| Total (n = 1 934) | SUDI referral centre (N = 1 593, 82.4%) | Hospital (n = 138, 7.1%) | Forensic medical institute (n = 203, 10.5%) | |
|---|---|---|---|---|
| Male gender | 1 098/1 919 (57.2) | 898/1 585 (56.7) | 86/137 (62.8) | 114/197 (57.9) |
| Age at the time of death (months), median (IQR) | 3.0 (1.0–6.0) | 3.0 (1.0–6.0) | 3.0 (1.0–5.0) | 4.0 (2.0–8.0) |
| Birth term, median (IQR) | 39.0 (37.0–40.0) | 39 (37.0–40.0) | 38 (36.0–40.0) | 38.0 (36.0–40.0) |
| Preterm | 334/1 647 (20.3) | 284/1 465 (19.4) | 27/95 (28.4) | 23/87 (26.4) |
| Low birth weight | 375/1 638 (22.9) | 323/1 461 (22.1) | 24/92 (26.1) | 28/85 (32.9) |
| Symptoms in the previous 72 h | 822/1 667 (49.3) | 734/1 480 (49.6) | 53/105 (50.5) | 35/82 (42.7) |
Note: Data are expressed as n/N (%).
Abbreviation: IQR, Interquartile range.
3.2. Postmortem Investigations
All the SUDI postmortem investigations are detailed in Table 2. The temporal trends observed for postmortem investigations from 2015 to 2024 are presented in Table 3. The completion rate of body temperature measurements decreased significantly from 82.7% to 68.7% (p < 0.001). The completion rate of chest X‐rays decreased significantly from 57.9% to 46.8% (p < 0.001). Conversely, a significant increase in fundus examination (from 9.3% to 31.6%, p < 0.001), laboratory tests (from 87.6% to 97.3%, p < 0.001), bacteriological tests (from 91.4% to 97.2%, p = 0.01), toxicological tests (from 79.0% to 91.0%, p < 0.001), and whole‐body imaging (from 75.2% to 87.3%, p < 0.001) was observed. Of the 1 934 cases, only 41 (2.1%) underwent exhaustive investigations in accordance with the French guidelines, with no significant change from 2015 to 2024. However, a median number of 10 (IQR: 8–11) postmortem investigations out of the 13 recommended were performed.
TABLE 2.
Postmortem investigations performed on SUDI cases in the whole cohort, and according to the site of admission.
| Total (n = 1 934) | SUDI referral centre (N = 1 593) | Hospital (n = 138) | Forensic institute (n = 203) | |
|---|---|---|---|---|
| Weight, size, and head circumference | 1 322/1 934 (68.4) | 1 224/1 593 (76.8) | 46/138 (33.3) | 52/203 (25.6) |
| Body temperature | 1 038/1 402 (74.0) | 967/1 282 (75.4) | 43/63 (68.3) | 28/57 (49.1) |
| Fundus examination | 347/1 607 (21.6) | 327/1 415 (23.1) | 3/99 (3.0) | 17/93 (18.3) |
| Laboratory tests | 1 601/1 722 (93.0) | 1 453/1 516 (95.8) | 87/106 (82.1) | 61/100 (61.0) |
| C‐Reactive Protein | 1 473/1 501 (98.1) | 1 355/1 380 (98.2) | 69/71 (97.2) | 49/50 (98.0) |
| Lumbar puncture | 1 490/1 692 (88.1) | 1 366/1 496 (91.3) | 68/102 (66.7) | 56/94 (59.6) |
| Bacteriological tests | 1 659/1 708 (97.1) | 1 497/1 515 (98.8) | 85/100 (85.0) | 77/93 (82.8) |
| Blood culture | 1 377/1 628 (84.6) | 1 256/1 472 (85.3) | 66/84 (78.6) | 55/72 (76.4) |
| Nasopharyngeal | 964/1 628 (59.2) | 879/1 472 (59.7) | 43/84 (51.2) | 42/72 (58.3) |
| Faces | 807/1 628 (49.6) | 748/1 472 (50.8) | 28/84 (33.3) | 31/72 (43.1) |
| Tracheal | 340/1 628 (20.9) | 306/1 472 (20.8) | 20/84 (23.8) | 14/72 (19.4) |
| Urine | 348/1 628 (21.4) | 316/1 472 (21.5) | 19/84 (22.6) | 13/72 (18.1) |
| Virological tests | 1 622/1 702 (95.3) | 1 459/1 505 (96.9) | 82/100 (82.0) | 81/97 (83.5) |
| Nasopharyngeal | 1 389/1 587 (87.5) | 1 267/1 432 (88.5) | 65/80 (81.2) | 57/75 (76.0) |
| Faces | 876/1587 (55.2) | 809/1 432 (56.5) | 32/80 (40.0) | 35/75 (46.7) |
| Tracheal | 233/1 587 (14.7) | 212/1 432 (14.8) | 15/80 (18.8) | 6/75 (8.0) |
| Toxicological tests | 1 337/1 584 (84.4) | 1 214/1 415 (85.8) | 53/87 (60.9) | 70/82 (85.4) |
| Blood | 1 196/1 275 (93.8) | 1 093/1 161 (94.1) | 44/51 (86.3) | 59/63 (93.7) |
| Urine | 246/1 275 (19.3) | 225/1 161 (19.4) | 9/51 (17.6) | 12/63 (19.0) |
| Gastric fluid | 175/1 275 (13.7) | 152/1 161 (13.1) | 9/51 (17.6) | 14/63 (22.2) |
| Hair | 114/1 275 (8.9) | 109/1 161 (9.4) | 1/51 (2.0) | 4/63 (6.3) |
| Bile | 42/1 275 (3.3) | 28/1 161 (2.4) | 0/51 (0.0) | 14/63 (22.2) |
| Skeletal X‐rays | 1 325/1 618 (81.9) | 1 209/1 440 (84.0) | 55/88 (62.5) | 61/90 (67.8) |
| Chest X‐rays | 690/1 512 (45.6) | 649/1 355 (47.9) | 24/82 (29.3) | 17/75 (22.7) |
| Brain imaging (CT scan or MRI) | 1 340/1 639 (81.8) | 1 195/1 440 (83.0) | 76/101 (75.2) | 69/98 (70.4) |
| Whole‐body imaging (CT scan or MRI) | 1 417/1 649 (85.9) | 1 278/1 456 (87.8) | 63/95 (66.3) | 76/98 (77.6) |
| Brain imaging and/or whole‐body imaging | 1 521/1 705 (89.2) | 1 358/1 494 (90.9) | 80/105 (76.2) | 83/106 (78.3) |
| Autopsy | 1 517/1 855 (81.8) | 1 272/1 548 (82.2) | 71/129 (55.0) | 174/178 (97.8) |
Note: Data are expressed as n/N (%).
Abbreviations: CT, Computed tomography; MRI, Magnetic resonance imaging.
TABLE 3.
Temporal trends observed for postmortem investigations performed on SUDI cases from 2015 to 2024.
| 2015* | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | Trend p | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Weight, size, and head circumference | 88 (72.7) | 139 (63.2) | 150 (65.8) | 167 (73.6) | 128 (71.9) | 128 (66.3) | 125 (69.4) | 161 (65.4) | 150 (71.1) | 86 (66.2) | 0.99 |
| Body temperature | 62 (82.7) | 115 (81.6) | 106 (80.9) | 130 (75.1) | 99 (78.0) | 109 (75.7) | 93 (66.9) | 133 (69.6) | 123 (67.6) | 68 (68.7) | < 0.001** |
| Fundus examination | 10 (9.3) | 23 (12.0) | 19 (10.3) | 32 (16.1) | 23 (15.1) | 37 (23.7) | 51 (32.7) | 57 (31.7) | 64 (35.2) | 31 (31.6) | < 0.001** |
| Laboratory tests | 99 (87.6) | 183 (91.0) | 174 (89.7) | 201 (94.8) | 146 (92.4) | 152 (92.7) | 153 (92.7) | 195 (95.6) | 191 (95.0) | 107 (97.3) | < 0.001** |
| C‐Reactive Protein | 84 (98.8) | 157 (98.7) | 152 (98.1) | 178 (96.7) | 140 (99.3) | 146 (98.0) | 146 (96.7) | 187 (98.9) | 184 (98.4) | 99 (98.0) | 0.94 |
| Lumbar puncture | 90 (81.1) | 169 (87.1) | 165 (84.6) | 180 (87.0) | 146 (92.4) | 148 (91.4) | 138 (85.2) | 175 (87.5) | 183 (92.9) | 96 (90.6) | < 0.01** |
| Bacteriological tests | 106 (91.4) | 193 (96.5) | 188 (97.9) | 202 (97.1) | 154 (98.1) | 158 (96.3) | 160 (97.0) | 198 (98.0) | 196 (99.5) | 104 (97.2) | 0.01** |
| Virological tests | 104 (91.2) | 183 (92.9) | 187 (95.4) | 202 (96.7) | 152 (96.2) | 154 (93.3) | 153 (95.0) | 194 (96.5) | 191 (97.9) | 102 (96.2) | 0.02** |
| Toxicological tests | 83 (79.0) | 138 (77.1) | 143 (78.6) | 157 (84.9) | 141 (91.6) | 129 (81.6) | 123 (86.0) | 158 (83.2) | 174 (92.6) | 91 (91.0) | < 0.001** |
| Skeletal X‐rays | 89 (79.5) | 154 (82.8) | 146 (79.3) | 175 (85.4) | 116 (80.6) | 119 (78.8) | 131 (84.0) | 162 (84.8) | 152 (80.0) | 81 (81.8) | 0.78 |
| Chest X‐rays | 62 (57.9) | 96 (54.9) | 87 (48.6) | 94 (48.2) | 52 (39.7) | 47 (36.4) | 65 (44.2) | 71 (38.8) | 72 (41.9) | 44 (46.8) | < 0.001** |
| Brain imaging (CT scan or MRI) | 95 (84.8) | 153 (80.5) | 155 (82.9) | 174 (82.9) | 123 (81.5) | 135 (82.8) | 132 (84.1) | 156 (83.4) | 147 (79.9) | 70 (71.4) | 0.16 |
| Whole‐body imaging (CT scan or MRI) | 85 (75.2) | 152 (78.8) | 157 (86.3) | 174 (86.1) | 135 (90.0) | 149 (92.0) | 135 (85.4) | 167 (86.5) | 174 (89.7) | 89 (87.3) | < 0.001** |
| Brain imaging and/or whole‐body imaging | 99 (86.8) | 171 (87.2) | 173 (89.2) | 184 (86.0) | 143 (91.7) | 156 (94.0) | 144 (88.3) | 175 (88.8) | 182 (91.9) | 94 (87.9) | 0.18 |
| Autopsy | 99 (81.8) | 172 (79.6) | 183 (83.2) | 178 (78.8) | 138 (79.3) | 152 (84.9) | 136 (79.5) | 194 (84.3) | 170 (84.2) | 95 (81.9) | 0.27 |
Note: Data are expressed as n (%). *Inclusions started in May 2015. ** Trend p < 0.05.
Abbreviations: CT, Computed tomography; MRI, Magnetic resonance imaging.
The reasons reported for not performing the postmortem investigations are presented in Table 4. The absence of laboratory tests was attributed to either failure to perform (27.3%) or not being requested (26.4%). The absence of bacteriological and virological tests was mainly due to not performing clinical examination (49.0% and 35.0%, respectively). The absence of autopsy was attributed to parental refusal (79.9%) or not being reported (12.1%).
TABLE 4.
Reasons for not performing postmortem investigations.
| Not performed | Outside the scope of the local protocol | Failure to perform | Not requested | No clinical examinations | Parental refusal | Police inquiry | Refusal or absence of a physician | Administrative issues | Other imaging modality | No information | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Weight, size, and head circumference* | 612/1 934 (31.6) | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR |
| Body temperature* | 364/1 402 (26.0) | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR |
| Fundus examination | 1 260/1 607 (78.4) | 907 (72.0) | 6 (0.5) | 243 (19.3) | — | 1 (0.1) | 10 (0.8) | 33 (2.6) | 1 (0.1) | — | 59 (4.7) |
| Laboratory tests | 121/1 722 (7.0) | 4 (3.3) | 33 (27.3) | 32 (26.4) | — | 4 (3.3) | 21 (17.4) | 14 (11.6) | — | — | 13 (10.8) |
| C‐Reactive Protein* | 28/1 501 (1.9) | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR |
| Lumbar puncture | 202/1 692 (11.9) | 8 (4.0) | 69 (34.2) | 60 (29.7) | — | 4 (2.0) | 19 (9.4) | 15 (7.4) | — | — | 27 (13.4) |
| Bacteriological tests | 49/1 708 (2.9) | 2 (4.1) | 1 (2.0) | 10 (20.4) | 24 (49.0) | 3 (6.1) | 2 (4.1) | 1 (2.0) | — | — | 6 (12.2) |
| Virological tests | 80/1 702 (4.7) | 16 (20.0) | 0 (0.0) | 20 (25.0) | 28 (35.0) | 3 (3.8) | 3 (3.8) | 0 (0.0) | — | — | 10 (12.5) |
| Toxicological tests | 247/1 584 (15.6) | 48 (19.4) | 41 (16.6) | 45 (18.2) | 40 (16.2) | 2 (0.8) | 3 (1.2) | 1 (0.4) | — | — | 67 (27.1) |
| Skeletal X‐rays | 293/1 618 (18.1) | 116 (39.6) | — | 58 (19.8) | — | 2 (0.7) | 9 (3.1) | 9 (3.1) | 1 (0.3) | 63 (21.5) | 35 (11.9) |
| Chest X‐rays | 822/1 512 (54.4) | 452 (55.0) | — | 88 (10.7) | — | 2 (0.2) | 13 (1.6) | 10 (1.2) | 1 (0.1) | 199 (24.2) | 57 (6.9) |
| Brain imaging | 299/1 639 (18.2) | 144 (48.2) | — | 72 (24.1) | — | 3 (1.0) | 11 (3.7) | 12 (4.0) | 2 (0.7) | 18 (6.0) | 37 (12.4) |
| Whole body imaging | 232/1 649 (14.1) | 94 (40.5) | — | 69 (29.7) | — | 3 (1.3) | 10 (4.3) | 12 (5.2) | 7 (3.0) | 2 (0.9) | 35 (15.1) |
| Autopsy | 338/1855 (18.2) | — | 41 (12.1) | — | 270 (79.9) | — | 6 (1.8) | 5 (1.5) | — | 16 (4.7) |
Note: Data are expressed as n (%).
Abbreviation: NR: Not requested in the SUDI registry.
The reasons for not performing anthropometric, body temperature or C‐reactive protein measurements are not recorded in the SUDI registry.
Of the 1 934 cases, 255 (13.2%) underwent minimal non‐invasive postmortem investigations, including fundus examination, toxicological tests, whole‐body imaging, skeletal X‐rays, and brain imaging. There was a significant increase in the rate of non‐invasive postmortem investigations performed, ranging from 6.6% in 2015 to 18.5% in 2024. The main reason reported for not performing the minimal non‐invasive postmortem investigations was that these examinations were not included in the local protocol. This applied to fundus examination (72.0%), toxicological tests (19.4%), whole‐body imaging (40.5%), skeletal X‐rays (39.6%), and brain imaging (48.2%).
3.3. Cause of Death
Of the 1 934 SUDI cases, 1 813 (93.4%) had been classified by the referral paediatrician: 217 cases (12.0%) in the SIDS group, 270 cases (14.9%) in the undetermined group and 1 326 cases (73.1%) in the explained group. The latter group comprises infectious causes (n = 306), suffocation/asphyxia (n = 561), cardiac causes (n = 88), metabolic causes (n = 16), toxic or traumatic causes (n = 65), other causes (n = 115) or multifactorial causes (n = 175). We found 121 (6.3%) cases could not be classified, and this was mainly due to the absence of police investigation results. Of the 41 cases that underwent exhaustive investigations in accordance with the French guidelines, 39 cases were classified: 14/217 (6.5%) in the SIDS group, 25/1326 (1.9%) in the explained group, and none were in the undetermined group. Of the 13 recommended postmortem investigations, a higher median number was performed in the SIDS group (11/13) than in the undetermined group (8/13).
4. Discussion
This study aimed to describe the postmortem investigations performed in 1 934 SUDI cases in daily practice in France, to assess their temporal trends from 2015 to 2024 and their compliance with the French guidelines. The minimum set of non‐invasive postmortem investigations required to detect non‐natural death was also studied.
First, we observed a high completion rate of laboratory tests, bacteriological tests, and virological tests, which were performed in 93.0%, 97.1%, and 95.3% of cases, respectively, during the study period. This could easily be explained by a good acceptance of percutaneous samples, which are accessible and cost‐effective for detecting virological and bacteriological infections and genetic or metabolic diseases [4, 17, 18, 19].
Second, we found significant changes in routine postmortem imaging, with a decrease in the use of chest X‐rays since 2015, progressively replaced by whole‐body CT scans. It had been well established that postmortem imaging plays an essential role in detecting traumatic injuries [20, 21]. The various imaging modalities have also been shown to be accurate in diagnosing the natural causes of death, particularly congenital abnormalities, with correct identification of the cause of death from 40% to 71% [22, 23, 24]. This change has been made possible through improved knowledge of postmortem imaging semiology, in both postmortem CT scans and magnetic resonance imaging, a better understanding of the value and limitations of imaging in SUDI cases and a greater availability of the devices [24]. Yet, a 2024 study has shown that skeletal radiography remains necessary, especially for detecting metaphyseal fractures, as its spatial resolution is better than that of postmortem CT scans [25]. These whole‐body X‐rays should continue to be performed, with radiographers being supervised if necessary to ensure images are acquired with the utmost rigour. Although postmortem CT scans are currently the main brain and whole‐body imaging modality, postmortem magnetic resonance imaging allows for better tissue differentiation, enabling the detection of subtle anatomical defects, particularly those affecting the heart or central nervous system [26, 27]. Access to postmortem magnetic resonance imaging remains difficult in France, but its widespread use could become a nationwide challenge in the coming decades.
Third, the completion rate of autopsies remained consistently high from 2015 to 2024, ranging from 81.8% to 81.9%. Unlike in other countries, autopsies are not mandatory in France, except in cases of judicial inquiry. As might be expected, the main reason given for not carrying out an autopsy was parental refusal (79.9%). This could be due to cultural or religious beliefs, difficulty accepting the damage caused to the infant's body, or doubts about the procedure's usefulness [28, 29]. Practitioners should continue their educational efforts to explain the purposes and the expected answers of an autopsy. They must bear in mind that postmortem imaging cannot fully replace an autopsy and that pathological investigations remain the most effective method for precisely determining the cause of death [4, 22, 30]. Improvements in practices could involve assessing the feasibility and accuracy of minimally invasive autopsies that combine postmortem imaging and imaging‐guided biopsies [31, 32]. These could be more acceptable to bereaved parents.
Finally, postmortem fundus examinations were the least frequently performed (21.6%) during the study period, but we found a significant increase from 9.3% in 2015 to 31.6% in 2024. An ophthalmological examination allows for the detection of signs of abusive head trauma, such as retinal haemorrhages or flares [33]. The difficulty in integrating a postmortem fundus examination into the local protocol highlights the need to coordinate all practitioners involved in SUDI management, including ophthalmologists [34].
Regarding the causes of death, the rate of SIDS causes (12.0%) was relatively low in comparison with the literature, likely due to an overestimation of asphyxia/suffocation causes [35, 36]. Providing further training for referral paediatricians on the application of international classifications, alongside the development of automated classification algorithms, could limit this bias in the future. As expected, the degree of compliance with the French guideline was higher in the unexplained or SIDS groups than in the undetermined group. This reinforces the evidence that only thorough postmortem investigations can provide strong arguments for SIDS and prevent deaths from being categorised as undetermined causes. Following the example of countries like the UK or most US states, the mandatory nature of postmortem investigations, including autopsies in every SUDI case, could be debated in France. The expected benefits in terms of public health policy regarding the understanding of early childhood mortality and the detection of non‐natural deaths should be balanced against the experiences of bereaved families.
These results were partially encouraging, as around three‐quarters of the recommended investigations were carried out in more than half of the cases. Yet, these findings were qualified by the fact that only 2.1% of infants included in the French SUDI registry underwent all the postmortem investigations recommended in the French guidelines. In addition, the most commonly reported reasons for not performing the postmortem investigations, particularly toxicological tests and whole‐body imaging, were that they were not included in the local protocol or requested. Although the French guidelines are well known among referral paediatricians, organisational issues still prevent their implementation in some centres. Local protocols should be improved through appropriate interventions once the local issues have been identified, with the systematic involvement of all relevant specialists, such as ophthalmologists, radiologists, and pathologists. In addition, an update to the French guidelines, which were developed almost 20 years ago, is necessary. The new French guidelines should consider the results of the present study regarding their suitability for real‐world conditions, particularly with regard to the limitations related to a lack of equipment or training, as well as technical advances, such as the use of microbiological PCR tests instead of viral and/or bacteriological cultures. The indications and procedures for taking complementary samples to detect metabolic disorders, including vitreous fluid samples, bile, and frozen organ samples, by combining with DNA‐sequencing techniques, could also be clarified, as they are not routinely carried out [37, 38].
The proportion of cases that underwent the minimum set of non‐invasive postmortem investigations to detect child abuse cases was particularly low, despite a significant increase over the study period. Inadequate or incomplete postmortem investigations increase the risk of missing inflicted traumatic injuries, with dramatic consequences for siblings and/or future children [39]. For this reason, in April 2023, a directive from the French Ministry of Health established that four examinations, including fundus examination, skeletal X‐rays, brain imaging, whole‐body imaging, and toxicological tests, are now mandatory. Any anomaly must be reported immediately to the judicial authorities [40].
4.1. Strengths and Limitations
This study provided updated and representative results on postmortem investigations performed as part of standard French practice. It was based on the collection of high‐quality data and the large number of cases included in the French SUDI registry. However, certain limitations must be acknowledged. Missing data relating to some postmortem investigations is mainly the result of police inquiries. In such cases, postmortem investigations may be incomplete due to funding issues related to the judicial system, or they may not be transmitted to the registry due to police confidentiality regulations.
5. Conclusion
Despite the significant increase in the frequency of certain investigations, particularly those designed to detect non‐natural deaths, only 2.1% of the SUDI cases were fully investigated following the French guidelines on the 2015–2024 period. The involvement of medical teams in referral centres is the cornerstone of postmortem investigations performed to determine the cause of death in SUDI cases. This study confirmed their commitment through an overall high completion rate from 2015 to 2024. Improvements in future practices should focus on understanding local issues in referral centres and forensic institutes and resolving them. In addition, the French guidelines should be updated appropriately. Joint efforts of referral centres, forensic institutes and judicial institutions would likely lead to more exhaustive postmortem investigations in SUDI cases. This would result in an improved diagnostic approach and subsequent tailored prevention strategies.
Author Contributions
Karine Levieux: conceptualization, methodology, visualization, writing – original draft. Christèle Gras‐Le Guen: writing – review and editing. Caroline Rambaud: writing – review and editing. Pauline Scherdel: conceptualization, methodology, software, formal analysis, writing – review and editing, supervision, visualization. Mathilde Ducloyer: conceptualization, methodology, writing – original draft, visualization.
Funding
The French national SUDI registry received funding from MSD Avenir, Sanofi Pasteur MSD, the French national public health agency (Santé Publique France), the national institute of health and medical research (Inserm), three French associations of parents, such as SA VIE, Naître et Vivre, and Les Rires d'Anna, and individual private donations.
Ethics Statement
The French SUDI registry was approved in 2015 by the French Data Protection Authority in clinical research (National Commission for Informatics and Freedoms, no. 915273) and by an independent ethics committee (Nantes Group for Ethics in the Health Sector, no. 2015‐01‐27). Parents or legal guardians of infants who died from SUDI signed a consent form in accordance with international recommendations.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: STROBE Statement—Checklist of items that should be included in reports of cross‐sectional studies (1).
Table S2: Complementary examinations recommended by the French National Health Authority (2).
Acknowledgements
We sincerely thank the families affected by SUDI who shared details of their personal lives with us and entrusted their child to us during this very difficult time in their lives. We also gratefully thank the medical and paramedical teams in the referral centres, including paediatricians, biologists, medical examiners, radiologists, ophthalmologists, nurses and psychologists who perform daily postmortem investigations, support parents and collect data for the French SUDI registry.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- 1. Krous H. F., Beckwith J. B., Byard R. W., et al., “Sudden Infant Death Syndrome and Unclassified Sudden Infant Deaths: A Definitional and Diagnostic Approach,” Pediatrics 114 (2004): 234–238. [DOI] [PubMed] [Google Scholar]
- 2. Duncan J. R., SIDS Sudden Infant and Early Childhood Death: The Past, the Present and the Future, 2018, ed. Byard R. W. (University of Adelaide Press, 2018), available from, https://www.adelaide.edu.au/press/titles/sids/. [PubMed] [Google Scholar]
- 3. Weber M. A., Ashworth M. T., Risdon R. A., Hartley J. C., Malone M., and Sebire N. J., “The Role of Post‐Mortem Investigations in Determining the Cause of Sudden Unexpected Death in Infancy,” Archives of Disease in Childhood 93 (2008): 1048–1053. [DOI] [PubMed] [Google Scholar]
- 4. Pries A. M., Van Der Gugten A. C., Moll H. A., et al., “Postmortem Diagnostics in Sudden Unexpected Death in Infants and Children: Use and Utility,” European Journal of Pediatrics 184 (2025): 223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Côté A., “Investigating Sudden Unexpected Death in Infancy and Early Childhood,” Paediatric Respiratory Reviews 11 (2010): 219–225. [DOI] [PubMed] [Google Scholar]
- 6. Garstang J., Ellis C., and Sidebotham P., “An Evidence‐Based Guide to the Investigation of Sudden Unexpected Death in Infancy,” Forensic Science, Medicine, and Pathology 11 (2015): 345–357. [DOI] [PubMed] [Google Scholar]
- 7. Arnestad M., Vege A., and Rognum T. O., “Evaluation of Diagnostic Tools Applied in the Examination of Sudden Unexpected Deaths in Infancy and Early Childhood,” Forensic Science International 125 (2002): 262–268. [DOI] [PubMed] [Google Scholar]
- 8. Pries A. M., Ruskamp J. M., Edelenbos E., et al., “A Systematic Approach to Evaluate Sudden Unexplained Death in Children,” Journal of Pediatrics 264 (2024): 113780. [DOI] [PubMed] [Google Scholar]
- 9. Bajanowski T., Vege A., Byard R. W., et al., “Sudden Infant Death Syndrome (SIDS)‐standardised Investigations and Classification: Recommendations,” Forensic Science International 165 (2007): 129–143. [DOI] [PubMed] [Google Scholar]
- 10. French Health Authority ‐ Guidelines for the investigation of sudden unexpected death in Infancy (under 2 years of age) (Haute Autorité de Santé , Prise en charge en cas de mort inattendue du nourrisson (moins de 2 ans)) [Internet], available from, https://www.has‐sante.fr/upload/docs/application/pdf/recommandations_mort_inattendue_nourrisson.pdf.
- 11. Bloch J., Denis P., and Jezewski‐Terra D., Steering Committee. National Institute for Public Health Surveillance: Unexpected Deaths Among Infants Under the Age of Two – National Survey 2007–2009 (Comité de pilotage. Institut National de Veille Sanitaire: Les morts inattendues du nourrisson de moins de 2 ans ‐ Enquête nationale 2007–2009, 2011), available from, https://www.santepubliquefrance.fr/sites/default/files/rdd/document/543878_inv9340.pdf. [Google Scholar]
- 12. Levieux K., Patural H., Harrewijn I., Hanf M., Gras Leguen C., and l'Association nationale des centres référents de la mort inattendue du nourrisson (ANCReMIN) , “Management of Sudden Unexpected Infant Death Syndrome (SUIDS) in Reference Centers in France in 2013,” Archives de Pediatrie: Organe Officiel de la Societe Francaise de Pediatrie 22 (2015): 360–367. [DOI] [PubMed] [Google Scholar]
- 13. Levieux K., Patural H., Harrewijn I., et al., “The French Prospective Multisite Registry on Sudden Unexpected Infant Death (OMIN): Rationale and Study Protocol,” BMJ Open 8 (2018): e020883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ducloyer M., de Visme S., Jarry B., et al., “The French Registry of Sudden Unexpected Death in Infancy (SUDI): A 7‐Year Review of Available Data,” European Journal of Pediatrics 183 (2024): 4991–5000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Von Elm E., Altman D. G., Egger M., Pocock S. J., Gøtzsche P. C., and Vandenbroucke J. P., “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies,” Journal of Clinical Epidemiology 61 (2008): 344–349. [DOI] [PubMed] [Google Scholar]
- 16. The 3rd International Congress on Sudden Infant and Child Death , Goldstein R. D., Blair P. S., et al., “Inconsistent Classification of Unexplained Sudden Deaths in Infants and Children Hinders Surveillance, Prevention and Research: Recommendations From the 3rd International Congress on Sudden Infant and Child Death,” Forensic Science, Medicine, and Pathology 15 (2019): 622–628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Neubauer J., Lecca M. R., Russo G., et al., “Post‐Mortem Whole‐Exome Analysis in a Large Sudden Infant Death Syndrome Cohort With a Focus on Cardiovascular and Metabolic Genetic Diseases,” Eur J Hum Genet EJHG 25 (2017): 404–409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Morris J. A., Harrison L. M., and Partridge S. M., “Postmortem Bacteriology: A Re‐Evaluation,” Journal of Clinical Pathology 59 (2006): 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Martin Perceval L., Scherdel P., Jarry B., de Visme S., Levieux K., and Gras‐Le G. C., “Sudden Unexpected Death in Infancy: Current Practices in Virological Investigations and Documentation in the French Registry,” Journal of Pediatrics 257 (2023): 113324. [DOI] [PubMed] [Google Scholar]
- 20. Kelly B. S., Van Rijn R. R., Bliss H., et al., “Post‐Mortem Imaging in Suspected Child Physical Abuse: A Systematic Review,” European Radiology (2026), https://link.springer.com/10.1007/s00330‐025‐12172‐1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Speelman A. C., Engel‐Hills P. C., Martin L. J., van Rijn R. R., and Offiah A. C., “Postmortem Computed Tomography Plus Forensic Autopsy for Determining the Cause of Death in Child Fatalities,” Pediatric Radiology 52 (2022): 2620–2629. [DOI] [PubMed] [Google Scholar]
- 22. Femia G., Langlois N., Raleigh J., et al., “Comparison of Conventional Autopsy With Post‐Mortem Magnetic Resonance, Computed Tomography in Determining the Cause of Unexplained Death,” Forensic Science, Medicine, and Pathology 17 (2021): 10–18. [DOI] [PubMed] [Google Scholar]
- 23. Shelmerdine S. C., Davendralingam N., Palm L., et al., “Diagnostic Accuracy of Postmortem CT of Children: A Retrospective Single‐Center Study,” AJR. American Journal of Roentgenology 212 (2019): 1335–1347. [DOI] [PubMed] [Google Scholar]
- 24. Proisy M., Marchand A. J., Loget P., et al., “Whole‐Body Post‐Mortem Computed Tomography Compared With Autopsy in the Investigation of Unexpected Death in Infants and Children,” European Radiology 23 (2013): 1711–1719. [DOI] [PubMed] [Google Scholar]
- 25. Shelmerdine S. C., Davendralingam N., Langan D., et al., “Post‐Mortem Skeletal Survey (PMSS) Versus Post‐Mortem Computed Tomography (PMCT) for the Detection of Corner Metaphyseal Lesions (CML) in Children,” European Radiology 34 (2024): 5561–5569. [DOI] [PubMed] [Google Scholar]
- 26. Arthurs O. J., Guy A., Thayyil S., et al., “Comparison of Diagnostic Performance for Perinatal and Paediatric Post‐Mortem Imaging: CT Versus MRI,” European Radiology 26 (2016): 2327–2336. [DOI] [PubMed] [Google Scholar]
- 27. Arthurs O. J., Thayyil S., Pauliah S. S., et al., “Diagnostic Accuracy and Limitations of Post‐Mortem MRI for Neurological Abnormalities in Fetuses and Children,” Clinical Radiology 70 (2015): 872–880. [DOI] [PubMed] [Google Scholar]
- 28. Sullivan J. and Monagle P., “Bereaved Parents' Perceptions of the Autopsy Examination of Their Child,” Pediatrics 127 (2011): e1013–e1020. [DOI] [PubMed] [Google Scholar]
- 29. Bunei M., Muturi P., Otiato F., et al., “Factors Influencing Acceptance of Post‐Mortem Examination of Children at a Tertiary Care Hospital in Nairobi,” Kenya. Annals Global Health 85 (2019): 95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Weber M. A., Pryce J. W., Ashworth M. T., Malone M., and Sebire N. J., “Histological Examination in Sudden Unexpected Death in Infancy: Evidence Base for Histological Sampling,” Journal of Clinical Pathology 65 (2012): 58–63. [DOI] [PubMed] [Google Scholar]
- 31. Lewis C., Latif Z., Hill M., et al., ““We Might Get a Lot More Families Who Will Agree”: Muslim and Jewish Perspectives on Less Invasive Perinatal and Paediatric Autopsy,” PLoS One 13 (2018): e0202023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Lewis C., Riddington M., Hill M., et al., “Availability of Less Invasive Prenatal, Perinatal and Paediatric Autopsy Will Improve Uptake Rates: A Mixed‐Methods Study With Bereaved Parents,” BJOG: An International Journal of Obstetrics & Gynaecology 126 (2019): 745–753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ducloyer J.‐B., Jossic F., Van Goethem V., et al., “Fatal Child Abuse Detected by Systematic Post‐Mortem Fundus Photograph in Sudden Death in Infancy,” International Journal of Legal Medicine 138 (2024): 1867–1872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Provost R., Ducloyer J.‐B., Rambaud C., et al., “Post‐Mortem Fundus Examination After Sudden Unexpected Death in Infancy (SUDI): Can We Do Better? A Study Based on the French SUDI Registry,” International Journal of Legal Medicine 140 (2025): 567–574. [DOI] [PubMed] [Google Scholar]
- 35. Parks S. E., Erck Lambert A. B., Hauck F. R., Cottengim C. R., Faulkner M., and Shapiro‐Mendoza C. K., “Explaining Sudden Unexpected Infant Deaths, 2011‐2017,” Pediatrics 147 (2021): e2020035873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Lowell G. S., Clark F., Ahadi R., and Quinlan K. P., “Using Sudden Unexpected Infant Death‐Case Registry Data to Drive Prevention,” Pediatrics 154 (2024): e2024067043J. [DOI] [PubMed] [Google Scholar]
- 37. The Royal college of pathologists, The Royal college of Paediatrics and Child Health , Sudden unexpected death in infancy and childhood, multi‐agency guidelines for care and investigation, 2016.
- 38. van Rijt W. J., Koolhaas G. D., Bekhof J., et al., “Inborn Errors of Metabolism That Cause Sudden Infant Death: A Systematic Review With Implications for Population Neonatal Screening Programmes,” Neonatology 109 (2016): 297–302. [DOI] [PubMed] [Google Scholar]
- 39. Fragkou K., Ketsekioulafis I., Tousia A., et al., “From Fragile Lives to Forensic Truth: Multimodal Forensic Approaches to Pediatric Homicide and Suspect Death,” Diagn Basel Switz 15 (2025): 1383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. [INSTRUCTION No. DGOS/R3/2023/44 of 19 April 2023 concerning the publication of the specifications for referral centres on sudden unexpected death in infancy ] , INSTRUCTION N° DGOS/R3/2023/44 du 19 avril 2023 relative à la diffusion du cahier des charges des centres de référence de la mort inattendue du nourrisson, 2023 N° DGOS/R3/2023/44 Apr 19.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: STROBE Statement—Checklist of items that should be included in reports of cross‐sectional studies (1).
Table S2: Complementary examinations recommended by the French National Health Authority (2).
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.