Forensic Science International 236 (2014) 16–21
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Experimental studies on the tensile properties of human umbilical cords Britta Tantius a, Markus A. Rothschild a, Markus Valter b, Joern Michael c, Sibylle Banaschak a,* a b c
Institute of Legal Medicine, University Hospital of Cologne, Melatenguertel 60/62, 50823 Cologne, Germany Department of Obstetrics and Gynecology, Cologne University, Kerpener Str. 34, 50931 Cologne, Germany Department of Orthopaedic and Trauma Surgery, University of Cologne, Kerpener Str. 62, 50937 Cologne, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 January 2013 Received in revised form 7 August 2013 Accepted 3 December 2013 Available online 16 December 2013
When tried in court, mothers accused of neonaticide may claim that the umbilical cord just broke during birth and the newborn child bled to death accordingly. To evaluate the possibility of a breakage of the umbilical cord is the goal of this work. Therefore 25 umbilical cords from neonates of both sexes born at term were stretched using an electrically operated material testing machine and the energy necessary to break them was measured. This experimental set-up equals a static strain, not a dynamic one. The maximum force endured (Fmax) ranged from 37.24 N to 150.04 N. The average force endured was 79.87 N with a standard deviation of 27.39. The elongation at break varied from 13.24% to a maximum of 119.93%. We found no relationship between the force endured and any of the following parameters: birth weight, pH of the venous umbilical blood, diameter of cord, free length under testing, duration of pregnancy or the mother’s age. We performed a literature research and tried to define the circumstances in which a break is more likely to occur, these being malformations, entanglement or disease, e.g. inflammation. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Umbilical cord Neonaticide Tensile strength Precipitate delivery
1. Introduction The classic case of mothers killing their newborn infants, called neonaticide, poses the examiner in a difficult position, as there are usually no witnesses available. The mothers often deny their pregnancy and finally give birth unobserved, often at home. When the dead infant is discovered, the mother’s account has to be compared with the autopsy findings. Mothers often claim that they were surprised by the sudden and unexpected delivery. It is reported in German textbooks [12] that some mothers stated that the umbilical cord ruptured during precipitate delivery. Due to this they reported that the child’s head hit the ground or that the child died due to the blood loss from the torn cord. In these cases, the forensic physician is asked to examine how the cord was disconnected (especially ruptured versus cut) and if it was torn, whether this could have happened by the weight of the newborn itself in the process of the spontaneous delivery.
* Corresponding author. Tel.: +49 0221 478 88327/88222. E-mail address: [email protected] (S. Banaschak). 0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.forsciint.2013.12.002
The resulting questions are: How much force is needed to result in a complete rupture of an umbilical cord? Is it possible that the umbilical cord will tear completely during precipitate delivery?
2. Materials and methods 2.1. Specimen 25 Umbilical cords were obtained from neonates of both sexes born at term (37 + 0 to 41 + 1 weeks gestation) after an uncomplicated pregnancy at the Department of Obstetrics and Gynecology, University Hospital Cologne. Institutional ethics approval was given and written consent was obtained from the mothers. The umbilical cords were cut at maximum available length (varied from 18 to 51 cm) and immediately stored in physiological saline solution in glass containers. This was meant to prevent alterations of mechanical properties by dehydration. Physiological saline solution does not contain all the specific elements of amniotic fluid, but has similar osmotic values to
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
17
Fig. 1. Scheme of Zwick/Ro¨ll Z050.
amniotic fluid at term [1], which seems the most important influence on mechanical properties. The umbilical cords were then either tested immediately or kept in a refrigerator at 4 8C and tested within 12 h after delivery at room temperature (18–22 8C). To make sure that no possible inflammation or malformation influenced the testing, a small disc was cut from either end and stored in 10% neutral buffered formalin for histopathological survey. The histological examination was performed using H&E staining. 2.2. Experimental set-up The machine used for testing the endured force was a Zwick/ Roell Z050, an electrically operated material testing machine with a moveable transverse arm (Fig. 1). The umbilical cord was attached to this at both ends by a specially designed fixator similar to a DD-belt closer (Figs. 2 and 3). This was meant to stop the cord from being squeezed flat, thus avoiding the possible bias of a changed cross section from circular to a flat oval. After reaching a preliminary force of 0.1 N, the specimen was then stretched at a constant rate of 50 mm/min until it broke. 3. Results The individual test readings for each umbilical cord are shown in Table 1. The maximum force endured (Fmax) ranged from 37.24 to 150.04 N (see Fig. 4). The average force endured was 79.87 N with a standard deviation of 27.39. The elongation at break varied from 13.24% to a maximum of 119.93%. The relationship of endured force to birth weight, pH of the venous umbilical blood, diameter of cord, free length under testing, duration of pregnancy or the mother’s age is shown in Table 2. These parameters were chosen because they were either proposed in previous studies [7,8,10] or because the authors suspected them to influence the mechanical properties of the umbilical cord. We could not find any relationship in these parameters. The torn ends of the umbilical cords were shredded, but laceration patterns differed widely. In all cases, the ruptured ends could be clearly identified in comparison with the ends previously cut by a scalpel. The free length under testing was a minimum of 5 cm. The maximum free length was 35 cm, with a mean value of 15.8 cm. Histopathological examination showed no pathological findings in any of the obtained specimens.
Fig. 2. Umbilical cord in fixator.
Fig. 3. DD-belt-like fixator.
4. Discussion 4.1. Neonaticide Neonaticide is defined in German jurisdiction as the killing of a newborn child by the mother sub partum or immediately post partum (up to 24 h). A retrospective analysis of 211 cases of neonaticide in Germany in the 1980s [2] draws the following picture: About 66% of the
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
18
160 140 120
[N]
100 80
Fmax [N]
60 40 20 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
[cases] Fig. 4. Maximum force endured.
Table 1 Parameters and results. No.
Mother’s age (y)
Infant’s weight (g)
Duration of pregnancy (w)
Arterial pH
Diameter of cord (mm)
Free length (cm)
Maximum strength (N)
Elongation at break (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
28 30 36 35 28 32 30 24 26 29 21 24 39 30 35 34 38 41 38 42 27 37 35 28 31
3830 3480 3600 3120 3800 3130 3190 3520 3780 4170 2910 3530 3280 2920 3810 2840 4160 3380 2630 3330 3400 3140 3680 3840 3460
38 + 5 39 + 0 37 + 0 40 + 2 39 + 3 40 + 1 38 + 3 39 + 2 40 + 3 39 + 6 39 + 6 39 + 5 38 + 0 38 + 6 39 + 1 38 + 5 41 + 0 37 + 4 37 + 3 39 + 0 41 + 1 38 + 6 39 + 1 40 + 0 41 + 0
7.28 7.37 7.31 7.29 7.32 7.41 7.44 7.19 7.37 7.41 7.33 7.26 7.38 7.19 7.29 7.32 7.37 7.36 7.46 7.32 7.20 7.30 7.36 7.33 7.29
10 10 10 10 10 10 14 11 9 11 8 9 12 10 11 10 14 12 7 12 10 9 15 9 11
10 10 23.9 10.1 10.1 6.9 27.7 15 25 12.8 10 20 15 25.5 5 19 10 10 10 10 35 15.1 20 20 20
60 126 87 98 50 107 66.64 49.61 89.91 76.23 65.28 56.7 67.68 63 73.81 111 37.24 96.48 90.16 120.96 68 72.09 54 59.94 150.04
34.11 37.63 14.55 18.67 44.85 52.83 13.24 45.11 15.53 119.93 41.29 25.1 32.62 16.44 55.11 34.59 41.74 30.52
a
a
41.67 a
46.07 32.26 30.44 19.17
In these two cases the testing machine did not recognize the break as such and therefore could not record the maximum elongation.
women were primiparae of an average age of 21.8 years. None of the women consulted a doctor on a regular basis; indeed in 83% no doctor was consulted either for diagnosis or prenatal check-up. The birth usually happened in the woman’s flat (81%), either sitting on the toilet (49%) or lying in bed (33%). 80.5% of the babies were not cared for after birth, a minority of 3.9% were cleaned of vernix caseosa and blood.
4.2. The umbilical cord
Table 2 Relationship of data.
Tensile strength of cord Tensile strength of cord Tensile strength of cord of cord (mm) Tensile strength of cord Tensile strength of cord of pregnancy (w) Tensile strength of cord
The further analysis of 163 autopsies of killed neonates shows that the mainly mature babies did not live long. If air was swallowed, it was only detected in the stomach. In German textbooks, it is generally stated that this indicates a lifespan of less than half an hour [3,4,12]. The umbilical cord was divided by sharp force in 70 of these cases, in 39 cases the end was shredded and in 31 cases the umbilical cord was intact.
Correlation coefficient
Relationship
(N)/birth weight (g) (N)/pH (N)/diameter
0.33 0.06 0.16
None probable None probable None probable
(N)/free length (cm) (N)/duration
0.12 0.11
None probable None probable
(N)/mother’s age (y)
0.29
None probable
The umbilical cord links the fetus to the placenta, where the fetal blood is oxygenated and enriched with nutrients. It contains three vessels: the umbilical vein, which carries the oxygenated blood from the placenta back to the fetus, and two umbilical arteries, which coil around the umbilical vein and contain blood low in oxygen. These vessels are surrounded by Wharton’s Jelly, a gelatinous connective tissue, consisting mainly of hyaluronic acid, in which collagenous and reticular fibers form a loose meshwork. This elastic but stable structure protects the umbilical vessels from kinking. The umbilical cord is covered by a single layer of cuboidal
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
amniotic epithelium and surrounded by amniotic fluid, which fills the amniotic sac and immerses the fetus [13]. 4.3. Why would it break? In 1948, Pommenich first asked, why an umbilical cord would break [5]. The umbilical cord could break either intrauterine, during birth or postnatal. In the very rare case of intrauterine breaking, a malformation or disease such as inflammation with subsequent adhesion of the cord to the amniotic sac, seems necessary. During birth, a certain amount of force needs to be endured. This strength could be impaired by malformations such as velamentous cord insertion or malformation of collagenous or reticular fibers. As well, there is the risk of cord entanglement. In these cases, the force is redirected and the umbilical cord is more likely to break. However, the umbilical cord does only break, if force or tension is applied. According to Sporrer (1995), there are two possible settings where the cord can break during delivery. One is a precipitate delivery whilst standing, the other getting up from a squatting position [6]. As stated above, another relevant setting is the spontaneous birth into toilets. In any case, the distance between birth canal and the surface on which the newborn comes to rest has to be greater than the length of the umbilical cord for the cord to be under tension. The main difficulty of breaking the cord manually is the slipperiness of the material. A tighter grip can be achieved via coiling the cord round the hands or by wrapping it with a rough material. The first option involves a minimum free length of approximately 55 cm. With the second option, it may be possible to find the used rough material, e.g. a cloth. 4.4. Literature review Table 3 summarizes results from other authors compared to our own data. The data had to be converted the original values to the SI-unit Newton (N). This was achieved by defining the acceleration as earth’s gravity. Thus, the mean breaking strength ranges from 49.03 N (Morris and Hunt [7]) to 73.55 N (Pommenich [5]). This study gives a slightly higher value of 79.87 N. These differences might be explained through an analysis of the differences in experimental setup. Pommenich (1948) used a static set-up by adding weights to the umbilical cord by means of artery clamps. He used whole afterbirths (cord with placenta) not older than three hours. In a second series, a weight of just 1 kg was attached to see if the cord broke after a certain free fall. This series was conducted because Pommenich could not measure force directly. The segment of free fall necessary to break the cord was apparently measured very roughly by looking at a tape measure attached next to it. The experimental set-up of a third series to examine the necessary velocity is full of possible inaccuracies: Velocity was determined by means of a string that was tightened manually by turning a reel. Time was measured with a second pendulum.
19
As a result, Pommenich stated that a weight of 1 kg and a free fall of 15 cm is enough to break the umbilical cord [5]. The samples used by Morris and Hunt (1966) were refrigerated for up to two days, then clamped in sandpaper-covered screw clamps and pulled manually by a spring balance. The longer remaining piece was pulled again until it broke. This was repeated up to three times. The authors stated that it is not difficult to break an umbilical cord by hand, but that the most serious difficulty is its slipperiness. They find no relation of tensile strength to length of umbilical cord. Umbilical vein varices are recognized as weak points and broken ends are described to have a characteristic appearance [7]. Spann and Englert (1967) unfortunately did not give any explanation as to the origin of the umbilical cords used, their storage or their fixation in the testing machine. They found that the longer the umbilical cord, the more work was necessary to break the cord and the less elongation could be found. The authors did not find any correlation between tensile strength and any of the following parameters: circumference of the cord, birth weight of the infant, duration of pregnancy of more than 31 weeks, number of coils and up to 15 days of storage [8]. Zink and Reinhardt (1969) used umbilical cords that were up to three days old (five were even older). They tried a few different settings, both static and dynamic, including the model of a precipitate delivery into a toilet. The placenta was held in a glass funnel, the umbilical cord led through its opening. This leads to a concave position of the placenta, as opposed to the natural convex position of the placenta attached to the uterine body. The elongation in the dynamic free-fall-test was measured via measuring tape. Weights were attached to umbilical cords by means of a piece of string, twined around the umbilical end (which in the opinion of the authors simply must have cut into the cord). In one series, they tried to break the cord by hand and found that it had to be done jerkily and using anti-slip-protection. The model of a precipitate delivery into a toilet (using a formalin-fixed stillbirth) is not consistent with a modern flush toilet. The toilet model they used was a simple frustum with a depth of 57 cm and an upper-rim-diameter of 33 cm manufactured of iron plate. They identified umbilical vein varices and thinner parts as weak points. There is also a clear mistake, possibly a typographical error: The authors give an elongation at break of 0.09–0.35%. This is more likely 9–35% [9]. Crichton (1973) wrote from an obstetrician’s point of view, trying to find out how much traction can be used in placental delivery. The specimens (cord and placenta) used came from mature infants after a spontaneous delivery, and the cord had not been under traction before. He found many ruptures occurred at placental insertion, maybe due to the fact that his placental specimens were hung up in hot-water bottles and thus in a concave position. As well, the examiner’s hand held the umbilical cord during testing, which probably influenced testing results. Crichton found no correlation of the tensile strength with either birth weight, weight of placenta, length of cord or the site of break [10]. Ghosh et al. (1984) used umbilical cords of 17 mature fetuses, which were stored in normal saline solution at 4 8C and tested
Table 3 Comparison of results (N = kg m s2 ! 1 kg equivalent 1 kp = 9.80665 N).
Pommenich Morris/Hunt Spann/Englert Zink/Reinhardt Crichton Ghosh/Ghosh/Gupta Sporrer Own results
Mean breaking strength (original unit)
Mean breaking strength (N)
Experimental set-up
7.5 kg 5 kg 5.7 kp The authors do not give a mean value 5.88 kg 6.73 kg 63.95 N 79.87 N
73.55 49.03 55.9
Static Dynamic Static Dynamic Static Static Static Static
57.66 66.0 63.95 79.87
20
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
within 8 h after delivery. Test specimens were only 5 cm long. As most other authors found several weak points along the length of the cord, testing only small pieces may not be representative of the whole umbilical cord. These samples were held in the machine in ‘‘appropriate’’ grips, though it is not further described, what these grips looked like. As this study was conducted in India, room temperature and humidity were probably higher than in our German climates [11]. Sporrer (1995) wrote from a technical point of view. His umbilical cord specimens were stored up to 14 h at 5 8C, then warmed to 20 8C room temperature. He used a testing machine similar to this work with a specially designed fixator, resembling a spiral. The static set-up was similar to our approach, but he applied more velocity, stretching the cords at 200 mm/min. There was also a dynamic set-up, where a weight of 3500 g was attached to the lower fixation via a 50 cm piece of string, so enabling a free fall of half a meter. All five specimens tested this way broke, but he reported a higher breaking strength than in the static approach. Thus the values given in the static set-up represent the lower limit of stress. Sporrer also looked at the predilection sites and stated that logically the cord would break at the minimum cross-section, which were supposed to be regularly found between middle and fetal third of the cord. Other possibly preferred sites of breakage are umbilical cord edemas and areas damaged during the delivery. He excluded the possibility of a rupture at the placental insertion, as this was in his view the part with the highest cross-section. He also excluded the rupture at the fetal navel, as the amniotic sheath here was stronger. As for getting up from a crouching position, he stated that the strength of the cord was enough to lift a child off the ground without breaking. To help reconstruct the course of unobserved delivery, he believed that due to the elasticity of the cord, the length of both remaining pieces could simply be added to get the original length [6]. 4.5. Evaluation of results According to Sporrer, the mean force of 79.87 N in our static setup represents the lower limit of the force an umbilical cord can endure [6]. The tensile strength in a dynamic set-up was found to be higher, which Sporrer explained via diminished exposure time to strain. Thus, in the case of a free fall, an umbilical cord is likely to be able to endure more than 80 N. All authors reported that if there is enough height in a free fall, the umbilical cord would rupture. In Sporrer’s free fall of 50 cm with a weight of 3.5 kg, 100% of the umbilical cords broke. Zink and Reinhardt [9] used a weight of 3.3 kg. 64% of their umbilical cords ruptured after a free fall of 5 cm, 88% ruptured after a fall of 10 cm. In all these set-ups, one has to bear in mind that the elongation of the cord must be added to the original length. For the cord to break, the available height therefore needs to be more than the elongated cord and the additional length of the infant (depending on the infant’s position). This is unlikely to happen in a squatting position on a western European toilet. It is however possible in a pit toilet, as used in other areas. 4.6. Site of rupture Crichton [10] and Zink and Reinhardt [9] find an increased number of ruptures at placental insertion. We believe that this is a bias formed by their experimental set-up. Both fixed the placenta into a concave position, resembling the position of the released placenta in the cervix. At the moment of birth, however, the placenta is still attached to the uterine body and thus in a convex position, diverting the applied force to a maximum diameter of the
placental surface. Our experimental set-up is not designed to verify this observation. Other authors describe various predilection sites, as stated above. Based on our study, we are not able to predict the site of rupture in the specimen. In an analysis of our photographic documentation, we did not find that sites of cord edema, or where the cord was of least diameter, or sites of umbilical vein varices were preferred sites of rupture. In summary, according to literature, a break is more likely to occur, if there is any sign of malformation or inflammation in the umbilical cord or in the uterus, or if there are any signs of umbilical cord entanglement around body parts. On the other hand, as varices and edema are present so often and complications so rare, their presence should not give way to assume a very muchheightened vulnerability. This leads again to the importance of a histopathological examination of umbilical cord and placenta. We should also bear in mind, that the delivery setting described must be one in which the umbilical cord is at least tightened. Because our experimental setting was static one, we cannot conclude from our data if a free fall of a newborn would break the umbilical cord. 4.7. Study limitations We tried to keep the time from delivery to testing as short as possible. Every specimen was transferred to normal saline solution, emulating amniotic fluid. If the specimen could not be tested right away, it was stored at 4 8C up to 12 h. Histological examination of the umbilical cords showed no autolysis and intact structural architecture. A bias caused by inadequate storage is therefore unlikely. The mechanical properties of the umbilical cord may vary over its length. We used the maximum available length, which means that a certain unknown length was cut when disconnecting the newborn, while the placenta was cut off with as little loss as possible. Further cord length was lost due to attachment of the cord at the fixator of the moveable transverse arm (see Fig. 3). Temperature very likely influences mechanical properties of the umbilical cords. Using the cooled specimens at the storage temperature of 4 8C would have made the material more brittle and less durable. In this experimental setting, we decided to go for room temperature, testing the umbilical cords at body temperature might have been another possible approach. The experimental set-up we used is static one. It could be argued, that a precipitate delivery is rather dynamic. This would be correct, if we imagine the parturient woman in a standing position and the baby in free fall. But, as stated above, most babies were delivered into a toilet or the parturient woman was lying in bed. For the cord to be under strain, the mother would have to get up. Thus, the umbilical cord is strained gradually. Under these circumstances, a static experimental approach seemed reasonable enough. It is a serious limitation of the study, that only 25 specimen could be obtained. Regarding the high variability in the force the umbilical cords could endure as well as in their elasticity and their site of rupture, a continuation of the study seems desirable to produce a more specific result. 5. Conclusions Breaking of umbilical cords may not occur often, but under forensic aspects, the single case is relevant. After all, in the individual case assessment, there is always the question: Accident or neonaticide? Regarding the results of this work, the mother’s claim of an umbilical cord rupture cannot be refuted. The variability of mechanical properties does not permit a general statement. The specific circumstances have to be reconstructed as detailed as possible. If available, not only the neonate but also additionally placenta and umbilical cord urgently have to be
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
examined, to check whether any malformations or inflammations might have influenced the course of events. Acknowledgements The authors are grateful to Prof. Dr. med. Mallmann and the entire staff of the Department of Obstetrics and Gynecology, University Hospital Cologne for their co-operation. The Department of Orthopaedic and Trauma Surgery kindly allowed us to use their testing machine. We also wish to thank Mr. Sporrer for providing us with a copy of his work. Dr. rer. nat. Inez Schulday and Benjamin Bromberger have proved themselves very generous in providing their expertise in physics. References [1] W. Schmidt, The Amniotic Fluid Compartment: The Fetal Habitat (Advances in Anatomy, Embryology and Cell Biology), Springer, Berlin, 1992, pp. 63–65. [2] H. Thomsen, M. Bauermeister, R. Wille, Zur Kindesto¨tung unter der Geburt. Eine Verbundstudie u¨ber die Jahre 1980–1981, Rechtsmed 2 (1992) 135–142.
21
[3] E. Rauch, B. Madea Kindesto¨tung, in: B. Madea, B. Brinkmann (Eds.), Handbuch gerichtliche Medizin, Springer, Berlin, 2003, pp. 933–934. [4] G. Adebahr, Kindsto¨tung, in: W. Schwerd (Ed.), Rechtsmedizin. Lehrbuch fu¨r Mediziner und Juristen, 5th ed., Deutscher A¨rzteverlag, Ko¨ln, 1992, p. 93. [5] C. Pommenich, Experimentelle Untersuchungen u¨ber die Zerreißfestigkeit der menschlichen Nabelschnur, (Dissertation), University of Bonn, 1948. [6] A. Sporrer, Rissfestigkeitsuntersuchungen an humanen Nabelschnu¨ren, Semesterarbeit, Ludwig-Maximilians-Universita¨t, Mu¨nchen, 1995. [7] J.F. Morris, A.C. Hunt, Breaking strength of the umbilical cord, J. Forensic Sci. 11 (1966) 43–49. [8] W. Spann, H.M. Englert, Experimentelle Untersuchungen der Zerreißfestigkeit der Nabelschnur,(Experimental studies on the laceration resistance of the umbilical cord), Dtsch. Z. Gesamt. Gerichtl. Med. 59 (1967) 196–200. [9] R. Zink, G. Reinhardt, Gerichtsmedizinische Untersuchungen zum Verhalten der Nabelschnur bei gewaltsamer Zerreißung,(Forensic studies on the behavior of the umbilical cord in violent rupture), Dtsch. Z. Gesamt. Gerichtl. Med. 66 (1969) 86–96. [10] J.L. Crichton, Tensile strength of the umbilical cord, Am. J. Obstet. Gynecol. 115 (1973) 77–80. [11] K.G. Ghosh, S.N. Ghosh, A.B. Gupta, Tensile properties of human umbilical cord, Indian J. Med. Res. 79 (1984) 538–541. [12] B. Madea, R. Dettmeyer, Kindesto¨tungen, in: B. Madea (Ed.), Praxis Rechtsmedizin, 2nd ed., Springer, Berlin, 2007, pp. 200–204. [13] D. Starck, Embryologie. Ein Lehrbuch auf allgemein biologischer Grundlage, 3rd ed., Thieme, Stuttgart, 1975, pp. 241–243.
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Experimental studies on the tensile properties of human umbilical cords Britta Tantius a, Markus A. Rothschild a, Markus Valter b, Joern Michael c, Sibylle Banaschak a,* a b c
Institute of Legal Medicine, University Hospital of Cologne, Melatenguertel 60/62, 50823 Cologne, Germany Department of Obstetrics and Gynecology, Cologne University, Kerpener Str. 34, 50931 Cologne, Germany Department of Orthopaedic and Trauma Surgery, University of Cologne, Kerpener Str. 62, 50937 Cologne, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 January 2013 Received in revised form 7 August 2013 Accepted 3 December 2013 Available online 16 December 2013
When tried in court, mothers accused of neonaticide may claim that the umbilical cord just broke during birth and the newborn child bled to death accordingly. To evaluate the possibility of a breakage of the umbilical cord is the goal of this work. Therefore 25 umbilical cords from neonates of both sexes born at term were stretched using an electrically operated material testing machine and the energy necessary to break them was measured. This experimental set-up equals a static strain, not a dynamic one. The maximum force endured (Fmax) ranged from 37.24 N to 150.04 N. The average force endured was 79.87 N with a standard deviation of 27.39. The elongation at break varied from 13.24% to a maximum of 119.93%. We found no relationship between the force endured and any of the following parameters: birth weight, pH of the venous umbilical blood, diameter of cord, free length under testing, duration of pregnancy or the mother’s age. We performed a literature research and tried to define the circumstances in which a break is more likely to occur, these being malformations, entanglement or disease, e.g. inflammation. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Umbilical cord Neonaticide Tensile strength Precipitate delivery
1. Introduction The classic case of mothers killing their newborn infants, called neonaticide, poses the examiner in a difficult position, as there are usually no witnesses available. The mothers often deny their pregnancy and finally give birth unobserved, often at home. When the dead infant is discovered, the mother’s account has to be compared with the autopsy findings. Mothers often claim that they were surprised by the sudden and unexpected delivery. It is reported in German textbooks [12] that some mothers stated that the umbilical cord ruptured during precipitate delivery. Due to this they reported that the child’s head hit the ground or that the child died due to the blood loss from the torn cord. In these cases, the forensic physician is asked to examine how the cord was disconnected (especially ruptured versus cut) and if it was torn, whether this could have happened by the weight of the newborn itself in the process of the spontaneous delivery.
* Corresponding author. Tel.: +49 0221 478 88327/88222. E-mail address: [email protected] (S. Banaschak). 0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.forsciint.2013.12.002
The resulting questions are: How much force is needed to result in a complete rupture of an umbilical cord? Is it possible that the umbilical cord will tear completely during precipitate delivery?
2. Materials and methods 2.1. Specimen 25 Umbilical cords were obtained from neonates of both sexes born at term (37 + 0 to 41 + 1 weeks gestation) after an uncomplicated pregnancy at the Department of Obstetrics and Gynecology, University Hospital Cologne. Institutional ethics approval was given and written consent was obtained from the mothers. The umbilical cords were cut at maximum available length (varied from 18 to 51 cm) and immediately stored in physiological saline solution in glass containers. This was meant to prevent alterations of mechanical properties by dehydration. Physiological saline solution does not contain all the specific elements of amniotic fluid, but has similar osmotic values to
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
17
Fig. 1. Scheme of Zwick/Ro¨ll Z050.
amniotic fluid at term [1], which seems the most important influence on mechanical properties. The umbilical cords were then either tested immediately or kept in a refrigerator at 4 8C and tested within 12 h after delivery at room temperature (18–22 8C). To make sure that no possible inflammation or malformation influenced the testing, a small disc was cut from either end and stored in 10% neutral buffered formalin for histopathological survey. The histological examination was performed using H&E staining. 2.2. Experimental set-up The machine used for testing the endured force was a Zwick/ Roell Z050, an electrically operated material testing machine with a moveable transverse arm (Fig. 1). The umbilical cord was attached to this at both ends by a specially designed fixator similar to a DD-belt closer (Figs. 2 and 3). This was meant to stop the cord from being squeezed flat, thus avoiding the possible bias of a changed cross section from circular to a flat oval. After reaching a preliminary force of 0.1 N, the specimen was then stretched at a constant rate of 50 mm/min until it broke. 3. Results The individual test readings for each umbilical cord are shown in Table 1. The maximum force endured (Fmax) ranged from 37.24 to 150.04 N (see Fig. 4). The average force endured was 79.87 N with a standard deviation of 27.39. The elongation at break varied from 13.24% to a maximum of 119.93%. The relationship of endured force to birth weight, pH of the venous umbilical blood, diameter of cord, free length under testing, duration of pregnancy or the mother’s age is shown in Table 2. These parameters were chosen because they were either proposed in previous studies [7,8,10] or because the authors suspected them to influence the mechanical properties of the umbilical cord. We could not find any relationship in these parameters. The torn ends of the umbilical cords were shredded, but laceration patterns differed widely. In all cases, the ruptured ends could be clearly identified in comparison with the ends previously cut by a scalpel. The free length under testing was a minimum of 5 cm. The maximum free length was 35 cm, with a mean value of 15.8 cm. Histopathological examination showed no pathological findings in any of the obtained specimens.
Fig. 2. Umbilical cord in fixator.
Fig. 3. DD-belt-like fixator.
4. Discussion 4.1. Neonaticide Neonaticide is defined in German jurisdiction as the killing of a newborn child by the mother sub partum or immediately post partum (up to 24 h). A retrospective analysis of 211 cases of neonaticide in Germany in the 1980s [2] draws the following picture: About 66% of the
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
18
160 140 120
[N]
100 80
Fmax [N]
60 40 20 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
[cases] Fig. 4. Maximum force endured.
Table 1 Parameters and results. No.
Mother’s age (y)
Infant’s weight (g)
Duration of pregnancy (w)
Arterial pH
Diameter of cord (mm)
Free length (cm)
Maximum strength (N)
Elongation at break (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
28 30 36 35 28 32 30 24 26 29 21 24 39 30 35 34 38 41 38 42 27 37 35 28 31
3830 3480 3600 3120 3800 3130 3190 3520 3780 4170 2910 3530 3280 2920 3810 2840 4160 3380 2630 3330 3400 3140 3680 3840 3460
38 + 5 39 + 0 37 + 0 40 + 2 39 + 3 40 + 1 38 + 3 39 + 2 40 + 3 39 + 6 39 + 6 39 + 5 38 + 0 38 + 6 39 + 1 38 + 5 41 + 0 37 + 4 37 + 3 39 + 0 41 + 1 38 + 6 39 + 1 40 + 0 41 + 0
7.28 7.37 7.31 7.29 7.32 7.41 7.44 7.19 7.37 7.41 7.33 7.26 7.38 7.19 7.29 7.32 7.37 7.36 7.46 7.32 7.20 7.30 7.36 7.33 7.29
10 10 10 10 10 10 14 11 9 11 8 9 12 10 11 10 14 12 7 12 10 9 15 9 11
10 10 23.9 10.1 10.1 6.9 27.7 15 25 12.8 10 20 15 25.5 5 19 10 10 10 10 35 15.1 20 20 20
60 126 87 98 50 107 66.64 49.61 89.91 76.23 65.28 56.7 67.68 63 73.81 111 37.24 96.48 90.16 120.96 68 72.09 54 59.94 150.04
34.11 37.63 14.55 18.67 44.85 52.83 13.24 45.11 15.53 119.93 41.29 25.1 32.62 16.44 55.11 34.59 41.74 30.52
a
a
41.67 a
46.07 32.26 30.44 19.17
In these two cases the testing machine did not recognize the break as such and therefore could not record the maximum elongation.
women were primiparae of an average age of 21.8 years. None of the women consulted a doctor on a regular basis; indeed in 83% no doctor was consulted either for diagnosis or prenatal check-up. The birth usually happened in the woman’s flat (81%), either sitting on the toilet (49%) or lying in bed (33%). 80.5% of the babies were not cared for after birth, a minority of 3.9% were cleaned of vernix caseosa and blood.
4.2. The umbilical cord
Table 2 Relationship of data.
Tensile strength of cord Tensile strength of cord Tensile strength of cord of cord (mm) Tensile strength of cord Tensile strength of cord of pregnancy (w) Tensile strength of cord
The further analysis of 163 autopsies of killed neonates shows that the mainly mature babies did not live long. If air was swallowed, it was only detected in the stomach. In German textbooks, it is generally stated that this indicates a lifespan of less than half an hour [3,4,12]. The umbilical cord was divided by sharp force in 70 of these cases, in 39 cases the end was shredded and in 31 cases the umbilical cord was intact.
Correlation coefficient
Relationship
(N)/birth weight (g) (N)/pH (N)/diameter
0.33 0.06 0.16
None probable None probable None probable
(N)/free length (cm) (N)/duration
0.12 0.11
None probable None probable
(N)/mother’s age (y)
0.29
None probable
The umbilical cord links the fetus to the placenta, where the fetal blood is oxygenated and enriched with nutrients. It contains three vessels: the umbilical vein, which carries the oxygenated blood from the placenta back to the fetus, and two umbilical arteries, which coil around the umbilical vein and contain blood low in oxygen. These vessels are surrounded by Wharton’s Jelly, a gelatinous connective tissue, consisting mainly of hyaluronic acid, in which collagenous and reticular fibers form a loose meshwork. This elastic but stable structure protects the umbilical vessels from kinking. The umbilical cord is covered by a single layer of cuboidal
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
amniotic epithelium and surrounded by amniotic fluid, which fills the amniotic sac and immerses the fetus [13]. 4.3. Why would it break? In 1948, Pommenich first asked, why an umbilical cord would break [5]. The umbilical cord could break either intrauterine, during birth or postnatal. In the very rare case of intrauterine breaking, a malformation or disease such as inflammation with subsequent adhesion of the cord to the amniotic sac, seems necessary. During birth, a certain amount of force needs to be endured. This strength could be impaired by malformations such as velamentous cord insertion or malformation of collagenous or reticular fibers. As well, there is the risk of cord entanglement. In these cases, the force is redirected and the umbilical cord is more likely to break. However, the umbilical cord does only break, if force or tension is applied. According to Sporrer (1995), there are two possible settings where the cord can break during delivery. One is a precipitate delivery whilst standing, the other getting up from a squatting position [6]. As stated above, another relevant setting is the spontaneous birth into toilets. In any case, the distance between birth canal and the surface on which the newborn comes to rest has to be greater than the length of the umbilical cord for the cord to be under tension. The main difficulty of breaking the cord manually is the slipperiness of the material. A tighter grip can be achieved via coiling the cord round the hands or by wrapping it with a rough material. The first option involves a minimum free length of approximately 55 cm. With the second option, it may be possible to find the used rough material, e.g. a cloth. 4.4. Literature review Table 3 summarizes results from other authors compared to our own data. The data had to be converted the original values to the SI-unit Newton (N). This was achieved by defining the acceleration as earth’s gravity. Thus, the mean breaking strength ranges from 49.03 N (Morris and Hunt [7]) to 73.55 N (Pommenich [5]). This study gives a slightly higher value of 79.87 N. These differences might be explained through an analysis of the differences in experimental setup. Pommenich (1948) used a static set-up by adding weights to the umbilical cord by means of artery clamps. He used whole afterbirths (cord with placenta) not older than three hours. In a second series, a weight of just 1 kg was attached to see if the cord broke after a certain free fall. This series was conducted because Pommenich could not measure force directly. The segment of free fall necessary to break the cord was apparently measured very roughly by looking at a tape measure attached next to it. The experimental set-up of a third series to examine the necessary velocity is full of possible inaccuracies: Velocity was determined by means of a string that was tightened manually by turning a reel. Time was measured with a second pendulum.
19
As a result, Pommenich stated that a weight of 1 kg and a free fall of 15 cm is enough to break the umbilical cord [5]. The samples used by Morris and Hunt (1966) were refrigerated for up to two days, then clamped in sandpaper-covered screw clamps and pulled manually by a spring balance. The longer remaining piece was pulled again until it broke. This was repeated up to three times. The authors stated that it is not difficult to break an umbilical cord by hand, but that the most serious difficulty is its slipperiness. They find no relation of tensile strength to length of umbilical cord. Umbilical vein varices are recognized as weak points and broken ends are described to have a characteristic appearance [7]. Spann and Englert (1967) unfortunately did not give any explanation as to the origin of the umbilical cords used, their storage or their fixation in the testing machine. They found that the longer the umbilical cord, the more work was necessary to break the cord and the less elongation could be found. The authors did not find any correlation between tensile strength and any of the following parameters: circumference of the cord, birth weight of the infant, duration of pregnancy of more than 31 weeks, number of coils and up to 15 days of storage [8]. Zink and Reinhardt (1969) used umbilical cords that were up to three days old (five were even older). They tried a few different settings, both static and dynamic, including the model of a precipitate delivery into a toilet. The placenta was held in a glass funnel, the umbilical cord led through its opening. This leads to a concave position of the placenta, as opposed to the natural convex position of the placenta attached to the uterine body. The elongation in the dynamic free-fall-test was measured via measuring tape. Weights were attached to umbilical cords by means of a piece of string, twined around the umbilical end (which in the opinion of the authors simply must have cut into the cord). In one series, they tried to break the cord by hand and found that it had to be done jerkily and using anti-slip-protection. The model of a precipitate delivery into a toilet (using a formalin-fixed stillbirth) is not consistent with a modern flush toilet. The toilet model they used was a simple frustum with a depth of 57 cm and an upper-rim-diameter of 33 cm manufactured of iron plate. They identified umbilical vein varices and thinner parts as weak points. There is also a clear mistake, possibly a typographical error: The authors give an elongation at break of 0.09–0.35%. This is more likely 9–35% [9]. Crichton (1973) wrote from an obstetrician’s point of view, trying to find out how much traction can be used in placental delivery. The specimens (cord and placenta) used came from mature infants after a spontaneous delivery, and the cord had not been under traction before. He found many ruptures occurred at placental insertion, maybe due to the fact that his placental specimens were hung up in hot-water bottles and thus in a concave position. As well, the examiner’s hand held the umbilical cord during testing, which probably influenced testing results. Crichton found no correlation of the tensile strength with either birth weight, weight of placenta, length of cord or the site of break [10]. Ghosh et al. (1984) used umbilical cords of 17 mature fetuses, which were stored in normal saline solution at 4 8C and tested
Table 3 Comparison of results (N = kg m s2 ! 1 kg equivalent 1 kp = 9.80665 N).
Pommenich Morris/Hunt Spann/Englert Zink/Reinhardt Crichton Ghosh/Ghosh/Gupta Sporrer Own results
Mean breaking strength (original unit)
Mean breaking strength (N)
Experimental set-up
7.5 kg 5 kg 5.7 kp The authors do not give a mean value 5.88 kg 6.73 kg 63.95 N 79.87 N
73.55 49.03 55.9
Static Dynamic Static Dynamic Static Static Static Static
57.66 66.0 63.95 79.87
20
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
within 8 h after delivery. Test specimens were only 5 cm long. As most other authors found several weak points along the length of the cord, testing only small pieces may not be representative of the whole umbilical cord. These samples were held in the machine in ‘‘appropriate’’ grips, though it is not further described, what these grips looked like. As this study was conducted in India, room temperature and humidity were probably higher than in our German climates [11]. Sporrer (1995) wrote from a technical point of view. His umbilical cord specimens were stored up to 14 h at 5 8C, then warmed to 20 8C room temperature. He used a testing machine similar to this work with a specially designed fixator, resembling a spiral. The static set-up was similar to our approach, but he applied more velocity, stretching the cords at 200 mm/min. There was also a dynamic set-up, where a weight of 3500 g was attached to the lower fixation via a 50 cm piece of string, so enabling a free fall of half a meter. All five specimens tested this way broke, but he reported a higher breaking strength than in the static approach. Thus the values given in the static set-up represent the lower limit of stress. Sporrer also looked at the predilection sites and stated that logically the cord would break at the minimum cross-section, which were supposed to be regularly found between middle and fetal third of the cord. Other possibly preferred sites of breakage are umbilical cord edemas and areas damaged during the delivery. He excluded the possibility of a rupture at the placental insertion, as this was in his view the part with the highest cross-section. He also excluded the rupture at the fetal navel, as the amniotic sheath here was stronger. As for getting up from a crouching position, he stated that the strength of the cord was enough to lift a child off the ground without breaking. To help reconstruct the course of unobserved delivery, he believed that due to the elasticity of the cord, the length of both remaining pieces could simply be added to get the original length [6]. 4.5. Evaluation of results According to Sporrer, the mean force of 79.87 N in our static setup represents the lower limit of the force an umbilical cord can endure [6]. The tensile strength in a dynamic set-up was found to be higher, which Sporrer explained via diminished exposure time to strain. Thus, in the case of a free fall, an umbilical cord is likely to be able to endure more than 80 N. All authors reported that if there is enough height in a free fall, the umbilical cord would rupture. In Sporrer’s free fall of 50 cm with a weight of 3.5 kg, 100% of the umbilical cords broke. Zink and Reinhardt [9] used a weight of 3.3 kg. 64% of their umbilical cords ruptured after a free fall of 5 cm, 88% ruptured after a fall of 10 cm. In all these set-ups, one has to bear in mind that the elongation of the cord must be added to the original length. For the cord to break, the available height therefore needs to be more than the elongated cord and the additional length of the infant (depending on the infant’s position). This is unlikely to happen in a squatting position on a western European toilet. It is however possible in a pit toilet, as used in other areas. 4.6. Site of rupture Crichton [10] and Zink and Reinhardt [9] find an increased number of ruptures at placental insertion. We believe that this is a bias formed by their experimental set-up. Both fixed the placenta into a concave position, resembling the position of the released placenta in the cervix. At the moment of birth, however, the placenta is still attached to the uterine body and thus in a convex position, diverting the applied force to a maximum diameter of the
placental surface. Our experimental set-up is not designed to verify this observation. Other authors describe various predilection sites, as stated above. Based on our study, we are not able to predict the site of rupture in the specimen. In an analysis of our photographic documentation, we did not find that sites of cord edema, or where the cord was of least diameter, or sites of umbilical vein varices were preferred sites of rupture. In summary, according to literature, a break is more likely to occur, if there is any sign of malformation or inflammation in the umbilical cord or in the uterus, or if there are any signs of umbilical cord entanglement around body parts. On the other hand, as varices and edema are present so often and complications so rare, their presence should not give way to assume a very muchheightened vulnerability. This leads again to the importance of a histopathological examination of umbilical cord and placenta. We should also bear in mind, that the delivery setting described must be one in which the umbilical cord is at least tightened. Because our experimental setting was static one, we cannot conclude from our data if a free fall of a newborn would break the umbilical cord. 4.7. Study limitations We tried to keep the time from delivery to testing as short as possible. Every specimen was transferred to normal saline solution, emulating amniotic fluid. If the specimen could not be tested right away, it was stored at 4 8C up to 12 h. Histological examination of the umbilical cords showed no autolysis and intact structural architecture. A bias caused by inadequate storage is therefore unlikely. The mechanical properties of the umbilical cord may vary over its length. We used the maximum available length, which means that a certain unknown length was cut when disconnecting the newborn, while the placenta was cut off with as little loss as possible. Further cord length was lost due to attachment of the cord at the fixator of the moveable transverse arm (see Fig. 3). Temperature very likely influences mechanical properties of the umbilical cords. Using the cooled specimens at the storage temperature of 4 8C would have made the material more brittle and less durable. In this experimental setting, we decided to go for room temperature, testing the umbilical cords at body temperature might have been another possible approach. The experimental set-up we used is static one. It could be argued, that a precipitate delivery is rather dynamic. This would be correct, if we imagine the parturient woman in a standing position and the baby in free fall. But, as stated above, most babies were delivered into a toilet or the parturient woman was lying in bed. For the cord to be under strain, the mother would have to get up. Thus, the umbilical cord is strained gradually. Under these circumstances, a static experimental approach seemed reasonable enough. It is a serious limitation of the study, that only 25 specimen could be obtained. Regarding the high variability in the force the umbilical cords could endure as well as in their elasticity and their site of rupture, a continuation of the study seems desirable to produce a more specific result. 5. Conclusions Breaking of umbilical cords may not occur often, but under forensic aspects, the single case is relevant. After all, in the individual case assessment, there is always the question: Accident or neonaticide? Regarding the results of this work, the mother’s claim of an umbilical cord rupture cannot be refuted. The variability of mechanical properties does not permit a general statement. The specific circumstances have to be reconstructed as detailed as possible. If available, not only the neonate but also additionally placenta and umbilical cord urgently have to be
B. Tantius et al. / Forensic Science International 236 (2014) 16–21
examined, to check whether any malformations or inflammations might have influenced the course of events. Acknowledgements The authors are grateful to Prof. Dr. med. Mallmann and the entire staff of the Department of Obstetrics and Gynecology, University Hospital Cologne for their co-operation. The Department of Orthopaedic and Trauma Surgery kindly allowed us to use their testing machine. We also wish to thank Mr. Sporrer for providing us with a copy of his work. Dr. rer. nat. Inez Schulday and Benjamin Bromberger have proved themselves very generous in providing their expertise in physics. References [1] W. Schmidt, The Amniotic Fluid Compartment: The Fetal Habitat (Advances in Anatomy, Embryology and Cell Biology), Springer, Berlin, 1992, pp. 63–65. [2] H. Thomsen, M. Bauermeister, R. Wille, Zur Kindesto¨tung unter der Geburt. Eine Verbundstudie u¨ber die Jahre 1980–1981, Rechtsmed 2 (1992) 135–142.
21
[3] E. Rauch, B. Madea Kindesto¨tung, in: B. Madea, B. Brinkmann (Eds.), Handbuch gerichtliche Medizin, Springer, Berlin, 2003, pp. 933–934. [4] G. Adebahr, Kindsto¨tung, in: W. Schwerd (Ed.), Rechtsmedizin. Lehrbuch fu¨r Mediziner und Juristen, 5th ed., Deutscher A¨rzteverlag, Ko¨ln, 1992, p. 93. [5] C. Pommenich, Experimentelle Untersuchungen u¨ber die Zerreißfestigkeit der menschlichen Nabelschnur, (Dissertation), University of Bonn, 1948. [6] A. Sporrer, Rissfestigkeitsuntersuchungen an humanen Nabelschnu¨ren, Semesterarbeit, Ludwig-Maximilians-Universita¨t, Mu¨nchen, 1995. [7] J.F. Morris, A.C. Hunt, Breaking strength of the umbilical cord, J. Forensic Sci. 11 (1966) 43–49. [8] W. Spann, H.M. Englert, Experimentelle Untersuchungen der Zerreißfestigkeit der Nabelschnur,(Experimental studies on the laceration resistance of the umbilical cord), Dtsch. Z. Gesamt. Gerichtl. Med. 59 (1967) 196–200. [9] R. Zink, G. Reinhardt, Gerichtsmedizinische Untersuchungen zum Verhalten der Nabelschnur bei gewaltsamer Zerreißung,(Forensic studies on the behavior of the umbilical cord in violent rupture), Dtsch. Z. Gesamt. Gerichtl. Med. 66 (1969) 86–96. [10] J.L. Crichton, Tensile strength of the umbilical cord, Am. J. Obstet. Gynecol. 115 (1973) 77–80. [11] K.G. Ghosh, S.N. Ghosh, A.B. Gupta, Tensile properties of human umbilical cord, Indian J. Med. Res. 79 (1984) 538–541. [12] B. Madea, R. Dettmeyer, Kindesto¨tungen, in: B. Madea (Ed.), Praxis Rechtsmedizin, 2nd ed., Springer, Berlin, 2007, pp. 200–204. [13] D. Starck, Embryologie. Ein Lehrbuch auf allgemein biologischer Grundlage, 3rd ed., Thieme, Stuttgart, 1975, pp. 241–243.
