Quantitative assessment of the diagnostic value of drowning signs based on a retrospective analysis of drowning cases

Full Text

BACKGROUND

Death by drowning is classified as violent death, that is why it always receives special attention from law enforcement and investigative bodies. Drownings account for approximately one-third of all cases of death due to mechanical asphyxia, highlighting the relevance of this issue [1, 2]. Determination of the cause of death during forensic autopsy of bodies recovered from water remains a challenging expert task, often an equation with many unknowns [3]. Unknown circumstances of the incident, “blurred” (literally and figuratively) morphological features, rapidly developing signs of putrefaction complicate diagnosis. In addition to the potential influence of internal factors (cardiovascular diseases, epilepsy, etc.), possible toxicological status should be considered. Alcohol intoxication is classified as an unfavorable exogenous factor that increases the risk of developing extreme conditions [4]. The effect of ethanol can contribute to atypical drowning cases characterized by the rapid onset of death with poor specific morphological features at autopsy. Alcohol in the blood often affects thanatogenesis, complicates the objective diagnosis of the causes of death, and can mask manifestations of drowning. A wide range of morphological features identified during the examination of bodies recovered from water is described in forensic medical practice [3, 5–16]. While previous studies were mainly focused on identifying new diagnostic criteria [3, 5–10], the current focuse is on assessing the informative value and diagnostic significance of both newly proposed and classical features of drowning [11–13, 15, 16]. For example, Stephenson et al. [11] evaluated the diagnostic significance of laboratory tests in establishing the fact of drowning, noting their limited reliability due to a high percentage of matches with control cases. After conducting a retrospective analysis of 331 forensic medical reports of drowning cases, Schneppe et al. [12] concluded that the examined signs lacked diagnostic value. The only exception was Krushevsky sign (external foam), which was observed exclusively in drowning victims. Lung emphysema, Sveshnikov sign, gastric mucosal damage, Paltauf spots, hemolytic staining of the intima of the aortic root, and other diagnostic signs of drowning require interpretation considering the time since death and resuscitation measures. Additionally, Schneppe et al. observed that pathomorphological changes typical of heart diseases were found in almost half of atypical drowning cases [12]. In their systematic review, Tyr et al. [13] analyzed postmortem pathological data used to diagnose drowning. After studying 86 original researches and 46 additional sources, the authors established that only the presence of foam in the respiratory tract, as well as emphysema and pulmonary edema with a postmortem interval of no more than 48 hours, can be considered as highly informative signs of drowning. Other signs, in their opinion, are not relevant or reliable enough for the diagnosis. Russian researchers also continue to study the specificity and informativeness of drowning signs [14–16]. Thus, this issue remains relevant and continues to provoke discussions in the forensic medical community.

AIM

To evaluate the diagnostic value of morphological features of drowning based on their frequency and statistical probability, as derived from archival data of the Moscow Region Bureau of Forensic Medical Examination for 2017–2019.

METHODS

Study Design

This was a retrospective, single-center, observational study.

Eligibility Criteria

Inclusion criteria:

• Cases of autopsy studies of drowning cases, where forensic histological, forensic chemical (with determination of blood and urine ethyl alcohol concentrations), and diatom test of internal organs and biological fluids (lungs, kidneys, contents of the spheroid sinus, blood [or washings] from the left ventricle of the heart) were conducted;
• Time of death not exceeding two days;
• Absence ofsigns ofputrefaction and

Non-inclusion criteria:

• Cases of studies of bodies recovered from water, where death resulted from mechanical trauma;
• Time of death and its circumstances unknown;
• Putrefaction of the body;
• Unidentified bodies (with unknown age).

Study Setting

Information was collected at the Moscow Region Bureau of Forensic Medical Examination (State budgetary institution of health care) based on the cooperation agreement No. 333/1-s/8 dated December 21, 2021. Archival expert reports of the Moscow Region Bureau of Forensic Medical Examination for 2017–2019 were used as research material. Forensic medical examinations, including diatom test (list of objects and features of their collection), were regulated by the Order dated May 12, 2010, No. 346n¹. “Guidelines for Conducting Forensic Medical Examination in Cases of Alcohol Poisoning” [17] were used to estimate the blood alcohol concentration.

Primary Outcome

Based on preliminary analysis and systematization of data from specialized sources, a program of signs was formed, including information on the circumstances of the incident, demographic characteristics of the drowned, results of forensic chemical analysis of blood and urine ethyl alcohol concentrations, data from diatom test, and morphological and general asphyxial signs of drowning.

Outcomes Registration

Data collection of autopsy and histological, forensic chemical, and diatom tests were conducted; their mathematical processing and grouping were performed. Outcomes were documented by creating a database in the Microsoft Excel® (Microsoft, USA) spreadsheets.

Ethics Approval

The study was approved by the Local Ethics Committee of Sechenov University (Minutes No. 13-25, dated June 5, 2025).

Statistical Analysis

Planned Sample Size

The sample size was not calculated in advance during planning and conducting the study.

Statistical Methods

Qualitative variables were presented as absolute and relative frequencies with 95% confidence intervals (CI). Comparison of qualitative characteristics was carried out using Pearson’s χ² test with subsequent pairwise comparison using the Bonferroni correction. Quantitative data distribution was assessed using the Shapiro–Wilk test. Since the distribution pattern in all cases did not correspond to normal pattern in at least one of the studied subgroups, the data are presented as Me [Q1; Q3], where Me is the median, Q1 and Q3 are the 1st and 3rd quartiles.

Statistical formal analysis was performed using IBM SPSS Statistics®, version 26.0 (IBM Corp., USA). Statistical significance was set at p < 0.05.

RESULTS

Sample Characteristics

The study included 179 cases of deaths from drowning, including 81.0% (n = 145) males (95% CI, 74.8–86.2) and 19.0% (n = 34) females (95% CI, 13.8–25.2). The age of individuals ranged from 6 months to 87 years, with a median value of 42 years [29; 57]. The distribution by age groups according to the WHO classification was as follows:

• 0–17 years, n = 13, 7.3% (95% CI, 4.1–8)
• 18–44 years, n = 86, 48.0% (95% CI, 40.8–3)
• 45–59 years, n = 43, 24.0% (95% CI, 18.2–7)
• 60–74 years, n = 34, 19.0% (95% CI, 13.8–2)
• 75–89 years, n = 3, 1.7% (95% CI, 0.5–4).

Primary Results

Details of the Incident

The peak of drowning deaths naturally occurs in the summer months:

• June, n = 31, 17.3% (95% CI, 12.3–4)
• July, n = 41, 22.9% (95% CI, 17.2–5)
• August, n = 42, 23.55% (95% CI, 17.7–1) (Fig. 1).

 

Fig. 1. Distribution of drowning cases by month.

 

Data on the date of the fatal incident were obtained from the circumstances of the case. Sunday was the most dangerous, with the highest number of drowning cases recorded [n = 35, 19.6% (95% CI, 9.9–20.2)], while Tuesday had the fewest [n = 15, 8.4% (95% CI, 5.0–13.1)]. On other days of the week, the frequency distribution was relatively even:

• Wednesday, n = 24, 13.4% (95% CI, 9.0–0)
• Thursday, n = 25, 14.0% (95% CI, 9.5–6)
• Friday, n = 27, 15.1% (95% CI, 10.4–9)
• Saturday, n = 26, 14.5% (95% CI, 9.9–2).

In 95 cases, bodies were found on the shore; in 61 cases, in natural bodies of water; in 18 cases, in pools; in 4 cases, in a bathtub; and in 1 case, in a well.

Macroscopic drowning signs

Persistent fine-bubble foam from/around the nose and mouth (Krushevsky sign) was recorded by experts in 33.5% of cases (95% CI, 26.9–40.7). Fine-bubble foam in the trachea and bronchi was identified in 135 cases (75.4%), with pink and white colors in 49.7% and 25.7% of cases respectively. “Wet emphysema” of the lungs (Emphysema aquosum) was observed by experts in 156 cases [87.2% (95% CI, 81.7–91.4)]. In 57.5% of cases (95% CI, 50.2–64.6), the lungs were described as heavy and enlarged due to fluid buildup, with fluid dripping from the surface. Rib impressions on the lung surface were found in 53.6% of cases (95% CI, 46.3–60.8). However, chest wall expansion and smoothing of the clavicle relief were not recorded by experts in the conclusions. Paltauf spots were the most frequently mentioned and identified in 89.95% of cases (95% CI, 84.9–93.7). Fluid buildup in the stomach and the first section of the small intestine (Fagerlund sign) was recorded in 37 cases, accounting for 20.75% of cases (95% CI, 15.2–27). Sveshnikov sign was identified in 84.45% of cases (95% CI, 78.5–89.1). In turn, the Rostoshinsky-Ulrich sign and Casper sign were recorded only in 11.2% (95% CI, 7.2–16.4) and 11.7% of cases (95% CI, 7.6–17.1) respectively. A rare finding was the gallbladder bed and wall edema recorded in 10 cases [5.6% (95% CI, 2.9–9.7)], whereas edema of the hepatoduodenal ligament was not recorded in any case. Furthermore, none of the forensic medical reports included in this study mentioned (no information about the presence/absence of) such signs as:

• Hemorrhages in the chest or neck muscles;
• Fluid in the tympanic cavities;
• Swelling of the epiglottis, mucosa of laryngopharynx, vocal folds;
• Casper sign;
• Three-layer stomach contents and tears in the stomach walls.

The frequency of general asphyxial signs is as follows:

• Livor mortis, 85.5% (95% CI, 79.8–90.1);
• Pitting edema of the face, 15.1% (95% CI, 10.4–20.9);
• Subconjunctival hemorrhages, 30.2% (95% CI, 23.8–37.2);
• Congestion of internal organs, 94.4% (95% CI, 90.3–97.1);
• Liquid blood in vessels and heart cavities, 88.3% (95% CI, 82.9–92.4);
• Congestion of the brain and its membranes, 80.4% (95% CI, 74.2–85.7);
• Cerebral edema, 48.6% (95% CI, 41.4–55.9);
• Tardieu spots, 27.4% (95% CI, 21.2–34.2);
• Splenic anemia, 12.8% (95% CI, 8.6–18.3); and
• Emptied bladder, 14.0% (95% CI, 9.5–19.6).

Microscopic signs

Diffuse venous and capillary congestion and pulmonary emphysema were noted in all cases of forensic diatom test, characterized by ruptured alveolar septa, acute vesicular emphysema, and dystelectasis. Cerebral edema was detected in 94.7% of cases (95% CI, 90.8–97.9), diffuse pulmonary hemorrhage, dystelectasis, and interstitial hemorrhages were recorded in 93.9% of cases (95% CI, 89.9–93.7). Homogeneous eosinophilic masses in the alveoli were recorded in 87.2% of cases (95% CI, 81.7–91.4), and pulmonary edema—64.2% (95% CI, 57.0–71).

Diatom test

Investigators provided water samples with diatoms (78%) and quartz-containing mineral particles (100%) in 22% of cases. In the diatom test of lung tissue samples, diatoms were detected in 41.3% of cases (95% CI, 34.3–48.6), and quartz-containing particles in 83.8% (95% CI, 77.9–88.6). In turn, the test was not conducted in 7 cases (3.9%). Diatoms were found in the fluid from the sphenoid sinus in 24.0% of cases (95% CI, 18.2–30.7), and quartz-containing mineral particles in 76.0% (95% CI, 69.3–81.8). In the diatom test of blood or washings from the left ventricle, plankton was detected in 4.5% of observations (95% CI, 2.1–8.3), quartz-containing mineral particles in 68.2% (95% CI, 61.1–74.6). Diatoms in kidney tissue were determined in only 2.2% of cases (95% CI, 0.8–5.2), whereas quartz-containing particles in 40.8% (95% CI, 33.8–48.1). The frequency of diatoms and quartz-containing mineral particles in tissues and biological fluids in the examined water samples was as follows:

• inlung tissue, 68.3% and2%;
• in thefluid from thesphenoid sinus, 30.5% and8%.

In the blood or washings of the left ventricle, the frequency of plankton and quartz-containing minarel particles was 6.8% and 69.0% when present in lung tissue, and 2.2% and 79.0% in water, respectively. In kidney tissue, the frequency of corresponding particles were 2.7% and 41.0% when present in lungs, and 1.7% and 40.6% in water, respectively. Considering that water samples are presented in only 22% of drowning cases, further study is needed with a focus on the correlation between the frequency of diatoms and quartz-containing mineral particles in organs and tissues and their presence (quantity, morphology, etc.) in water samples on an expanded sample.

Forensic chemical analysis

In 69.8% of cases, the results of forensic chemical examinations of blood and urine ethanol concentrations were positive: 125 drowned individuals were intoxicated. Ethanol was found in the blood and urine of 72.4% of males (95% CI, 64.7–79.2) and 58.8% of females (95% CI, 42.1–74.1), without statistical significance (p = 0.120). In the age group of 0–17 years, ethanol was found in 7.7% of cases; in the group 18–44 years, 79.1%; 45–59 years, 76.7%; 60–74 years, 64.7%; individuals over 74 years, 33.3% of cases. The highest average concentrations of ethanol were found in drowned individuals aged 18–44 years, and in the group 45–59 years (Table 1). The highest proportion of drowned in a state of severe, heavy, and lethal intoxication was also noted in the age group 18–44 years (Table 2).

 

Table 1. Alcohol concentration in blood and urine
Alcohol concentration	Age, years					p
	0–17	18–44	45–59	60–74	75–89
In blood, ‰	0.001.2.3 [0.00; 0.00]	2.401 [0.44; 3.50]	2.202 [0.42; 3.30]	1.803 [0.00; 2.80]	0.00 [0.00; 0.3]	< 0.001
In urine, ‰	0.001.2.3 [0.00; 0.00]	3.101 [0.44; 4.00]	2.802 [0.45; 3.90]	2.303 [0.00; 3.90]	0.00 [0.00; 0.00]	< 0.001
Note. Results are presented as Me [Q1; Q3], where Me is the median, Q1 and Q3 are the 1st and 3rd quartiles. 1,2,3 indicate the presence of statistically significant differences in pairwise comparison of the corresponding groups (p < 0.05 with Bonferroni correction).

 

Table 2. Degree of alcohol intoxication in age groups
Stage of intoxication	Age, years					p
	0–17	18–44	45–59	60–74	75–89
Not detected, %	92.31.2.3 (69.3–99.2)	20.91 (13.4–30.4)	23.32 (12.6–37.3)	35.33 (20.9–52.0)	66.7 (17.7–96.1)	0.001
Mild, %	0.0 (0.0–0.0)	9.3 (4.5–16.8)	4.7 (1.0–14.1)	2.9 (0.3–12.9)	33.3 (3.9–82.3)
Mean, %	0.0 (0.0–0.0)	7.0 (3.0–13.8)	14.0 (6.0–26.5)	11.8 (4.1–25.6)	0.0 (0.0–0.0)
Severe, %	0.0 (0.0–0.0)	17.4 (10.6–26.5)	11.6 (4.6–23.6)	8.8 (2.5–27.5)	0.0 (0.0–0.0)
Very severe, %	7.7 (0.8–30.7)	37.2 (27.6–47.7)	46.5 (32.2–61.2)	41.2 (25.9–57.9)	0.0 (0.0–0.0)
Fatal (>5‰), %	0.0 (0.0–0.0)	8.1 (3.7–15.3)	0.0 (0.0–0.0)	0.0 (0.0–0.0)	0.0 (0.0–0.0)
Note. The results are presented as the relative number of cases with a 95% confidence interval. 1,2,3 presence of statistically significant differences in pairwise comparison of the corresponding groups (p < 0.05 with Bonferroni correction).

 

We analyzed the dependence of the frequency of certain signs of drowning on the stage of intoxication. The selection of signs was based on ranked frequency (results of the previous stage), and on data from specialized resources on the possible influence of ethyl alcohol on their manifestations. Statistical analysis showed no significant differences (Table 3). In addition, no significant differences were found between the lung weight (right and left) depending on the stage of alcohol intoxication (Table 4).

 

Table 3. Degree of alcohol intoxication with various of drowning signs
Sign	Stage of intoxication
	Not detected	Mild	Mean	Severe	Very severe	Fatal
Krushevsky sign, %	38.3 (26.8–50.9)	10.0 (4.3–19.5)	10.0 (4.3–19.5)	8.3 (3.3–17.3)	28.3 (18.1–40.6)	5.0 (1.4–12.7)
Fine-bubbled foam in the tracheal and bronchial lumen:
pink	31.5 (22.5–41.6)	9.0 (4.3–16.2)	10.1 (5.1–17.6)	12.4 (6.7–20.4)	32.6 (23.5–42.8)	4.5 (1.5–10.3)
white	23.9 (13.4–37.6)	2.2 (0.2–9.7)	6.5 (1.9–16.4)	19.6 (10.1–32.7)	45.7 (31.9–59.9)	2.2 (0.2–9.7)
No funding, %	34.1 (21.4–48.8)	6.8 (2.0–17.1)	9.1 (3.1–20.2)	6.8 (2.0–17.1)	38.6 (25.3–53.4)	4.5 (1.0–13.8)
Pulmonary edema	25.2 (17.6–34.2)	7.8 (3.7–14.1)	11.7 (6.5–18.9)	14.6 (8.8–22.3)	35.9 (27.1–45.5)	4.9 (1.9–10.3)
Emphysema aquosum, %	30.1 (23.3–37.6)	5.1 (2.5–9.4)	8.3 (4.8–13.4)	14.1 (9.3–20.2)	37.8 (30.5–45.6)	4.5 (2.0–8.6)
Paltauf spots, %	31.1 (24.3–38.5)	6.8 (3.7–11.5)	8.1 (4.6–13)	11.8 (7.5–17.5)	38.5 (31.3–46.2)	3.7 (1.6–7.5)
Fagerlund sign, %	35.1 (21.3–51.2)	2.7 (0.3–11.9)	5.4 (1.1–16.2)	21.6 (10.8–36.7)	27.0 (14.8–42.7)	8.1 (2.3–20.1)
Sveshnikov sign, %	26.5 (19.9–33.9)	6 (3.0–10.6)	9.9 (5.9–15.5)	14.6 (9.6–20.8)	38.4 (30.9–46.3)	4.6 (2.1–8.9)
Quartz-containing mineral particles in the kidney, %	12.5 (1.4–45.4)	0.0 (0.0–0.0)	12.5 (1.4–45.4)	37.5 (11.9–70.5)	25.0 (5.6–59.2)	12.5 (1.4–45.4)
Quartz-containing mineral particles in the blood/washes from the left ventricle, %	27.0 (19.8–35.4)	6.6 (3.1–12.0)	7.4 (3.7–13.0)	14.8 (9.3–21.8)	40.2 (31.8–49)	4.1 (1.6–8.7)
Note. The results are presented as the relative number of cases with a 95% CI.

 

Table 4. Lung weight depending on the degree of alcohol intoxication
Stage of intoxication	Lung weight, grams
	Right	Left
Not detected	619 [518.25; 697]	542 [473.5; 630]
Mild	580 [513; 708]	535 [431.75; 644]
Mean	724 [576.5; 864]	641 [494.5; 807.5]
Severe	662 [486; 804]	562 [466; 710]
Very severe	670 [540; 786]	583 [482; 730]
Fatal (>5‰)	684 [590; 982]	564 [482; 730]
Note. The results are presented as Me [Q1; Q3], where Me is the median, Q1 and Q3 are the 1st and 3rd quartiles.

 

Pilot project of a computer program for assessing the probability of drowning

Based on the frequency rating of drowning signs (Table 5), we developed a pilot project for a computer program to assess its probability. Python supplement can be useful in forensic practice, helping to expedite expert decisions in determining the cause of death. One of the conditions was ease of use: data entry without complex structures. Tkinter, a standard tool for creating graphical interfaces in Python, was chosen to implement the user interface. To enhance its appearance, CustomTkinter was used, expanding the capabilities of standard tkinter elements with modern and attractive widgets.

 

Table 5. Frequency rating of drowning signs
Sign	n	%	(95% CI), %
Paltauf spots	161	89.9	(84.9–93.7)
Emphysema aquosum	156	87.2	(81.7–91.4)
Sveshnikov sign	151	84.4	(78.5–89.1)
Quartz-containing particles in the fluid from the sphenoid sinus	136	76.0	(69.3–81.8)
Fine-bubble foam in the trachea and bronchial lumen	135	75.4	(62.1–89.5)
Quartz-containing particles in left ventricle	122	68.2	(61.1–74.6)
Pulmonary edema (acute pulmonary hyperinflation)	103	57.5	(50.2–64.6)
Rib impressions on the lungs	96	53.6	(46.3–60.8)
Quartz-containing mineral particles in the kidney	73	40.8	(33.8–48.1)
Krushevsky sign	60	33.5	(26.9–40.7)
Fagerlund sign	37	20.7	(15.2–27)
Rostoschinsky-Ulrich sign	20	11.2	(7.2–16.4)

 

The program implements an algorithm for calculating the probability of drowning, based on a system of weighting diagnostic signs. For creating the application, we used signs that, according to the results obtained, were most frequently identified in forensic medical examination of drowning cases. Several key signs known as specific for drowning are included (see Table 5):

• Krushevsky sign
• Paltauf spots
• plankton in the heart and lungs
• Sveshnikov sign
• rib impressions on the lungs
• pulmonary edema and

We calculated the statistical probability for each sign, reflecting its contribution to the overall probability of drowning:

• Krushevsky sign: 6.95%
• Paltauf spots:18.66%
• Sveshnikov sign: 17.50%
• quartz-containing mineral particles in the sphenoid sinus: 15.76%
• rib impressions on the lungs: 11.12%
• pulmonary edema: 11.94%
• pulmonary emphysema: 18.08%
• total: 100%.

The calculation logic is implemented in the clicked function (Fig. 2), which processes the pressing of the Calculate button. It collects the values of all variables (1 or 0) and calculates the final probability of drowning based on the data on signs.

 

Fig. 2. Implementation of the clicked function for calculating the probability of drowning.

 

DISCUSSION

This study showed that males predominated among those who drowned, with most deaths in the most economically active age groups—young and middle-aged individuals according to the WHO classification. Similar demographic trends were identified in some Russian and international works [18–20] and in the work by Matvienko et al. [15], dedicated to the analysis of drowning cases in Minsk. According to Real et al. [20], the majority of drowned individuals were also men of working age (16–65 years, 61%). Our results indicate that drownings are an alcohol-attributable deaths. More than two-thirds of drowned individuals, men and women, were in a state of alcohol intoxication, indicating a direct correlation between alcohol intoxication, regardless of its concentration, and the risk of drowning. The consumption of ethanol before swimming is a predisposing factor. Even a low blood alcohol content can lead to behavioral disorders, increasing the danger when entering the water, especially in unfamiliar places. Kovalev et al. [21] noted that among cases of violent death, the highest frequency of alcohol intoxication is observed in mechanical asphyxia, including drowning, which accounts for 54.8%–59.8% from 2011 to 2016. At the same time, in deaths from mechanical trauma and exposure to extreme temperatures, the frequency of intoxication is 36.95%–43.1% and 49.3%–52.7%, respectively. Similar results were obtained during an extended statistical analysis: the average number of drowning cases in the Russian Federation, accompanied with alcohol intoxication, from 2013 to 2022 was 61.6%±2.7% [2]. Being a poison, ethanol has a systemic effect on the human body, affecting all organs and tissues. Its action goes through the excitation phase followed by the development of inhibition in the central nervous system [4]. The narcotic and toxic effects of ethyl alcohol have been thoroughly studied and described in the works of the scientific school of Pigolkin [22]. The impact of ethanol on the central nervous system and cognitive functions contributes to uncoordinated movement, deterioration of swimming skills, reduced ability to self-rescue, and increases the risk of falling into water and drowning death. According to the work by Pajunen et al. [23], more than 60% of drowned individuals not related to the use of water transport [International Classification of Diseases 10th Revision (ICD-10) W65–74] had a blood alcohol concentration of ≥50 mg/dL; the number of watercraft-related drownings (ICD-10 V90, V92) was even higher.

In this study, definitions such as type of drowning, wet drowning, dry drowning, etc., were not used. This is due to the fact that the type of drowning is not included when the forensic medical diagnosis is formulated and coded. Moreover, drowning classified into aspirational and asphyxial is debatable and currently criticized. This aspect is thoroughly examined in the dissertation work by Chumakova [3], who notes that outdated and misleading terminology is still widely used. Based on the definition itself, modifiers like pre-drowning, dry and wet drowning, and delayed and secondary drowning, cannot exist, yet used in medical work.

Many morphological features of drowning allow to determine the fact of death. Their diagnostic value has been discussed repeatedly. Currently, it is customary to distinguish characteristic signs used for diagnosing drowning and general asphyxial signs. However, conducted analytical review shows that their diagnostic value and frequency are contradictory.

According to the results, a rank-ordered frequency list of signs was formed (see Table 5). The most frequently observed were Paltauf spots. Being analogous to Tardieu spots, Paltauf spots have a different morphological picture due to the hemolyzing and diluting effect of water [8]. Tyr et al. [13] assess this sign as a component of Emphysema aquosum, which also includes:

• foam in the tracheal and bronchial lumen
• enlarged lungs with rib impressions
• certain histological signs.

In this study, we analyzed these signs separately, and their frequency varied (see Table 5). For instance, foam in the foam in the tracheal and bronchial lumen was recorded in 75.4% of drowning cases, whereas Krushevsky sign was in 33.5%. Similar figures are provided by Schneppe et al. [12]—73.3% and 35.6%, respectively. At the same time, Tyr et al. [13] noted that foam in the tracheal and bronchial lumen occurs in only 4% to 25% of cases and is not a characteristic diagnostic sign of drowning, as it can result from narcotic and alcohol intoxication, and heart failure. Moreover, Schneppe et al. [12] found foam in the tracheal and bronchial lumen in 14.3% of non-drowning cases. It is also known that performing cardiopulmonary resuscitation results in false-negative outcomes for this sign.

In this study, the second most frequent sign of drowning is Emphysema aquosum (87.2%). According to Tyr et al. [13] and Schneppe et al. [12], its determination rate was 65%–89% and 94.9%, respectively. However, Tyr and Schneppe emphasize the need for a critical approach to interpreting this sign, focusing on the correct assessment of morphological changes in the lungs and recommending the use of digital morphometry combined with artificial intelligence technologies. According to our data, Sveshnikov sign was observed in 87.2% of cases, with quartz-containing particles in the contents of the sphenoid sinus in 84.4% of cases, and diatom plankton in only 24.0%. According to Tyr et al. [13], fluid in the sinus was detected in 90%–100% of cases, whereas Matvienko et al. [15] and Schneppe et al. [12] noted this sign in 71.4% and 86.3% of cases, respectively. We support the authors’ opinion that its interpretation should include a quantitative assessment — the volume of fluid, which significantly enhances the diagnostic value of this sign. In the study by Schneppe et al. [12], the average volume of the contents was 1.6 ml. The frequency of pulmonary edema (large and heavy lungs filled with water) in our study was 57.5%. Tyr et al. [13], analyzing this sign, note its high subjectivity, similarly criticizing proposals to use threshold values of the total lung weight. In our opinion, a promising index is the drowning index—the ratio of the weight of the spleen to the lungs [13], although its threshold values remain a subject of discussion. Poulain et al. [24] found that lung weight in drowning cases significantly increases in cases of alcohol intoxication and xenobiotic use. In our study, no significant correlation between lung weight and the stage of alcohol intoxication was found (see Table 4). As one option for objectively assessing lung edema, it is proposed to use their ultrasound density or computed tomography data. The diagnostic capabilities of modern medical visualization methods are demonstrated in the work of Chumakova [3]. The results of the study showed that detecting diatom plankton in the kidney, fluid from the sphenoid sinus, and the left ventricle cavity are extremely rare, whereas quartz-containing mineral particles were observed more frequently. For a long period, the diatom test was considered the most objective and evidential. However, according to differrent resources, the accuracy of the test is contradictory for various reasons: the destruction of diatom during extraction, the diatom poverty of water bodies, and technical contamination of samples [18]. When analyzing publications on diatom plankton in drowning, the variability of positive results is noteworthy: from 14% [18] to 99.9% [19]. The absence of diatoms in the blood and internal organs may be related not only to the diatom characteristics of the body of water but also to the presence of ethanol in the blood of drowning individuals. Alcohol intoxication influences the mechanism of drowning:

• Reduction or omission of certain phases of drowning;
• Predominance ofedema overemphysema—reducing thepossibility ofdiatoms penetrating the

The results obtained refute this assumption: no statistically significant differences were found between the detection of diatoms and quartz-containing mineral particles in the left ventricular cavity (wash/blood) and alcohol intoxication (see Table 4). Poulain et al. [24] also concluded that there is no connection between the results of the diatom test and the presence of alcohol or drugs in the blood of drowned individuals. However, the research sample in this work is extremely small (only 19 drowning cases), and Poulain et al. [24] used the lung to detect diatoms. However, assessing the significance of the presented results of the positive diatom test frequency is limited, due to the lack of data on the diatom characteristics of the body of water.

We noted Fagerlund sign in only 20.7% of drowning cases. In the work by Matvienko et al. [15], it was observed even less frequently, in 10.71% of cases. In international sources, the diagnostic value of this sign is questioned due to considerable discrepancies in data on the amount of ingested fluid [13]. Meanwhile, the authors note that three-layering of stomach contents and longitudinal tears of the gastric mucosa may be more specific drowning signs.

In the analyzed medical documents, we did not find information on intramuscular hemorrhage of the neck and chest, hemolytic staining of the intima of aortic root, or the use of any additional research methods, except for diatom test.

Study Limitations

The sample size necessary to achieve the required statistical power of the results was not calculated when planning and conducting the study. Therefore, the sample obtained during the study cannot be considered sufficiently representative. This does not allow extrapolating the results obtained and their interpretation to the general population outside the study.

We calculated the average volume of fluid as 2.3 ml (95% CI, 0.5%–5.2%), but, due to the small sample size and the absence of volume indication of the detected fluid in some medical conclusions, this figure is not presented in the study results. Furthermore, the wide CI indicates a high degree of uncertainty in parameter estimation.

Moreover, that examining the correlation between spleen and lung weight was not included in the objectives of the study. We studied the relationship between the lung weight and the stage of alcohol intoxication. The hypothesis of a possible correlation is based on well-known data about the toxic effects of ethyl alcohol on internal organs, such as toxic pulmonary edema [22, 25].

CONCLUSION

Young and middle-aged men are most often among the drowned individuals, which confirms the socio-economic problem. A direct correlation has been established between the presence of ethanol in the blood and urine, regardless of concentration, and the risk of death from drowning. In the analyzed reports of the Moscow Region Bureau of Forensic Medical Examination, diagnostic signs of drowning found in ≥50% of cases include: Paltauf spots, emphysema and pulmonary edema, Sveshnikov sign, foam in the tracheal and bronchial lumen, rib impressions on the lung surface, quartz-containing mineral particles in the sphenoid sinus and in the left ventricular cavity in combination with general asphyxial signs. None of the identified drowning signs provided absolute diagnostic accuracy. Only a comprehensive approach and combined assessment of diagnostic features, when properly interpreted, can ensure the formation of an evidence-based expert conclusion on the cause of death individuals recovered from water. The analysis showed no statistically significant differences between the frequency of drowning signs, lung weight, and stage of alcohol intoxication. Developing an ideal diagnostic model for drowning remains a promising direction for future research.

ADDITIONAL INFORMATION

Author contributions: D.P. Kalashnikov, M.A. Kislov, D.A. Syrova: investigation, data curation, writing—original draft; N.V. Polukhin: formal analysis, writing—original draft, visualization; S.N. Zakharov: software, visualization; G.V. Zolotenkova: conceptualization, methodology, data curation, writing—original draft, writing—review & editing. All the authors approved the version of the manuscript to be published and agreed to be accountable for all aspects of the work, ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Ethics approval: The study was approved by the Local Ethics Committee of Sechenov University (Minutes No. 13-25, dated June 5, 2025).

Funding sources: No funding.

Disclosure of interests: The authors have no relationships, activities, or interests for the last three years related to for-profit or not-for-profit third parties whose interests may be affected by the content of the article.

Statement of originality: No previously obtained or published material (text, images, or data) was used in this study or article.

Data availability statement: The editorial policy regarding data sharing does not apply to this work.

Generative AI: No generative artificial intelligence technologies were used to prepare this article.

Provenance and peer-review: This article was submitted unsolicited and reviewed following the standard procedure. The peer-review process involved two members of the Editorial Board and the in-house science editor.

 

¹ Order of the Ministry of Health and Social Development of the Russian Federation No. 346n dated May 12, 2010: “On the Approval of the Procedure for the Organization and Production of Forensic Medical Examinations in State Forensic Institutions of the Russian Federation. (document not in force). Available at: https://base.garant.ru/12177987/ Accessed on: February 25, 2025.

About the authors

Denis P. Kalashnikov

City Clinical Hospital named after V.V. Veresaev

Author for correspondence.
Email: den_kalashnikov@mail.ru
ORCID iD: 0000-0002-4608-3038
SPIN-code: 8358-7004

MD

Russian Federation, Moscow

Nikita V. Polukhin

Moscow University “Synergy”

Email: nikitasketch@gmail.com
ORCID iD: 0000-0002-9540-5793
SPIN-code: 8047-9245

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Sviatoslav N. Zakharov

Sechenov First Moscow State Medical University (Sechenov University)

Email: zakharov.swyatoslaw@yandex.ru
ORCID iD: 0000-0003-0107-9649
SPIN-code: 7201-9898

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Maksim A. Kislov

The Russian National Research Medical University named after N.I. Pirogov

Email: kislov_ma@rsmu.ru
ORCID iD: 0000-0002-9303-7640
SPIN-code: 3620-8930

MD, Dr. Sci. (Medicine), Professor

Russian Federation, Moscow

Daria A. Syrova

Sechenov First Moscow State Medical University (Sechenov University)

Email: syrova_d_a@student.sechenov.ru
ORCID iD: 0009-0004-0761-6251
Russian Federation, Moscow

Galina V. Zolotenkova

Sechenov First Moscow State Medical University (Sechenov University)

Email: zolotenkova.galina@bk.ru
ORCID iD: 0000-0003-1764-2213
SPIN-code: 1685-1802

MD, Dr. Sci. (Medicine), Professor

Russian Federation, Moscow

References

Supplementary files