Forensic Sci Med Pathol (2016) 12:33–39 DOI 10.1007/s12024-015-9730-4
ORIGINAL ARTICLE
Sudden infant death syndrome: no significant expression of heat-shock proteins (HSP27, HSP70) Elke Doberentz1 • Sarah Fu¨hring1 • Burkhard Madea1
Accepted: 16 November 2015 / Published online: 12 December 2015 Ó Springer Science+Business Media New York 2015
Abstract Purpose In industrialized countries, sudden infant death is the most common cause of death in young children. Although prone sleeping position is a well-known risk factor, hyperthermia might also be important. Pathognomonic findings of premortem hyperthermia do not exist. During stress, including thermal effects, heat-shock protein (HSP) expression increases. This study investigated hyperthermia as a contributing or pathogenic factor for sudden infant death syndrome (SIDS). Methods Immunohistochemical staining for HSP27 and HSP70 in the kidney, heart, and lung from 120 SIDS cases was examined. Results HSP70 immunostaining was negative in kidney, heart, and lung tissues in all cases and in tissues from the control group. HSP27 staining was positive in the kidney from one case, and was positive in the lungs (respiratory epithelia in 27 % of cases; vascular endothelia in 19 % of cases) and was negative in the heart. In the control group HSP27 was positive in 8 % of renal tubular tissues and in 29 % of renal vascular endothelia. Staining for HSP27 in lung tissues was positive in respiratory epithelia in 8 % of cases and for vascular endothelia in 29 %, whereas tissues from the heart were positive in only 4 %. Conclusion The hypothesis of hyperthermia being a pathogenic factor for SIDS was not supported by immunohistochemical visualization of HSP70 or HSP27.
& Burkhard Madea [email protected] 1
Institute of Forensic Medicine, University of Bonn, Stiftsplatz 12, 53111 Bonn, Germany
Keywords SIDS Hyperthermia Body temperature Stress Immunohistochemistry Heat-shock proteins
Introduction Sudden infant death syndrome (SIDS) involves the sudden and unexpected death of apparently healthy infants. Despite extensive postmortem examinations of SIDS infants, the cause of death is poorly understood. In industrialized countries, sudden infant death is the most common cause of death among children during their first years of life. Several risk factors for SIDS have been identified, and the most well known is prone sleeping position [1–8]. Besides prone sleeping position, hyperthermia is also considered a possible risk factor. Thermal stress may increase the risk of SIDS, for example from high ambient temperatures, warm clothing or bedding, and increased body temperature during infectious diseases [2, 9–11]. Interestingly, sleeping position itself may predispose to hyperthermia [11–18]. In a study by Pfeifer on 107 of 142 postmortem SIDS babies, rectal body temperatures over 40 °C were measured [19, 20]. According to Kleemann et al. [11] the role of hyperthermia in sudden infant death has been discussed for many years. As a result of structured interviews with parents of SIDS victims (n = 140), signs of profuse sweating (moist head and hair and damp clothing or bedding) were noted at the death scenes in 35.7 % of cases compared to controls in the same geographical region. Sweat on the head and sweaty clothing has a significant association with the risk of SIDS. According to these studies, the pathophysiological basis for hyperthermia remains unclear. It might result from infections, over insulation, excessive clothing in high
123
34
environmental temperatures, covering of an infant’s head or immature central thermoregulatory centers. A case–control study on sleep environment risk factors for SIDS revealed that duvets double the risk for SIDS compared with infants using a sleeping bag only, or a very light cotton blanket [8]. Another case–control study confirmed that SIDS cases were characterized by excess thermal insulation for the given room temperatures compared with matched controls [17]. Overheating was independent to prone sleeping position associated with an increased risk for sudden infant death. In summary, there is evidence from a number of scene findings as well as from case–control studies that some infants experience thermal stress and that infant hyperthermia may have a role in the etiology of SIDS. However, according to Byard [2] it is likely that hyperthermia might be a risk modifier in infants who sleep in a prone position. This conclusion is based on the findings of Ponsonby et al. [17] who showed that SIDS cases are significantly more likely to be overdressed for the room temperature at the death scene and who were sleeping in the prone position compared with control infants. In cases of stress, including thermal effects, the expression of heat-shock proteins (HSP) may increase as a response to cellular stress. HSP are rapidly expressed with both endogenous and exogenous cellular stress. They support protein folding of newly synthesized proteins and are important for the stabilization and degradation of incorrectly folded or damaged proteins. They are divided into subgroups according to their molecular weights and can be stained for immunohistochemically [21–33]. HSPs are also expressed in renal tissue in cases of lethal hypothermia [34]. In a study group of 100 cases of death due to hypothermia, renal glomerular expression of HSP70 was present in 80 % of cases, and renal tubular expression was found in 89 % of cases compared with the control group (n = 50) with short and long agonal periods where no renal glomerular expression was found in 84 % of cases and no renal tubular expression was observed in 66 % of all cases [34]. HSPs are also rapidly expressed in various organ systems in cases of death due to fire [35]. HSP27 has a molecular weight of 27 kDa and belongs to a group of small HSP. Small HSP are present in the cytosol of most cells, even in the absence of stress factors; under heat-shock conditions their expression is increased dramatically. Their main purpose is the stabilization of microfilaments. HSP70 has a molecular weight of 70–78 kDa. At least eight subgroups have been identified in humans. HSP70 is expressed in the cytosol, nucleus, mitochondria, and endoplasmic reticulum of the cell and accounts for 1–2 % of the total cellular proteins. Their main purpose is protein folding, protein translocation, and cytoprotection.
123
Forensic Sci Med Pathol (2016) 12:33–39
The objective of the current study was to determine whether increased HSP27 or HSP70 expression is associated with premortem hyperthermia in SIDS. Immunohistochemical staining for HSP70 and HSP27 was performed on the hearts, lungs, and especially the kidneys, because they have a rapid HSP reaction after thermal stress [35].
Materials and methods The study group consisted of 120 SIDS cases (46 female and 74 male individuals). The diagnoses of SIDS as a cause of death were based upon the criteria of the San Diego Consensus Definition from 2004 [36]. Most cases belonged to category Ib. The minimum age was 7 days and the maximum was 344 days (mean age 125 days). The control group consisted of 29 cases of death in children due to unnatural causes (13 female and 16 male individuals) with a minimum age of 4 days and a maximum age of 354 days. During each autopsy, organ samples were obtained and stored in formaldehyde solution until further examination. For this study, the tissues were first dehydrated using an increasing sequence of alcohol solutions. Subsequently they were embedded in paraffin. The tissue blocks were then cut into 2–4 lm-thin slices and transferred to microscope slides. After drying in an incubator, the slides were dewaxed and rehydrated. Immunohistochemical staining for HSP70 and HSP27 were carried out according to the labeled streptavidin–biotin method with specific primary antibodies (Novo CastraÒ HSP27 mouse monoclonal antibody; Novo CastraÒ HSP70 mouse monoclonal antibody), a biotinylated secondary antibody and a labeled streptavidin complex. Staining of the specimens was completed with a chromogen substrate (Dako AEC Substrate Chromogen Code K3464) and hemalum for staining of the nuclei. Positive and negative controls were used in each staining process to verify the findings and to confirm antibody specificity. For examination of the stained samples, 30 visual fields of each specimen were semiquantitatively evaluated at 4009 magnification. Heart, lungs, and especially kidneys were chosen for investigation, because kidneys have a rapid HSP expression after thermal stress. Evaluation of the heart with epicardium, myocytes disci intercalares, perinuclear spaces, vessels, and fibrocytes; lungs with peripheral and central bronchial tubes, vessels (endothelia, lumina), interalveolar septa, pleura, peribronchial glands, peribronchial connective tissues, ciliated epithelia as well as the central respiratory tracts with bronchial tubes, ciliated epithelia, peribronchial glands, vessels (endothelia, lumina), cartilages, glandular tissues, and lymphatic tissues; and the kidneys with renal tube glomeruli, vessels, capsules, and connective tissues were used for immunohistochemical analysis, which was graded
Forensic Sci Med Pathol (2016) 12:33–39
semiquantitatively using the four-degree scale reported by Preuss et al. [34]. The numbers of positive red stained structures/cells observed in each visual field were estimated as a percentage of all investigated structures/cells. For each analyzed slice and structure, a mean value from all 30 visual fields was calculated and graduated according to the pattern in Table 1.
Results Tables 2 and 3 show the immunohistochemical analysis results for HSP70 and HSP27 in the cellular structures of each organ. Representative images of the immunohistochemical analysis for HSP27 and HSP70 stainings are shown in Figs. 1 and 2.
35
cells were stained positive for HSP27, respectively (Fig. 1a–d; Table 2). Control group In all 29 cases, the HE-stained sections showed no remarkable abnormal structures in tissues. In 4 % of cases, heart endothelium cells were stained positive for HSP27. In 8 and 4 % of cases, kidney tubular cells and vascular endothelium cells were stained positive for HSP27, respectively. In 8 and 29 % of cases, lung respiratory epithelium cells and vascular endothelium cells were stained positive for HSP27, respectively (Table 2).
HSP70 immunohistochemical analysis
HSP27 immunohistochemical analysis
Study group
Study group
In all 120 cases, HE-stained sections showed no remarkable abnormal structures in tissues. In all cases, no HSP70 stained structures were observed in the tissues from hearts, lungs, or kidneys (Fig. 2a, b; Table 3).
In all 120 cases, the hematoxylin and eosin (HE)-stained sections showed no remarkable abnormal structures in tissues. No stained structures were observed in tissue samples from hearts. In 1 % of cases, kidney tubular cells were stained positive for HSP27. In 27 and 19 % of cases, lung respiratory epithelium cells and vascular endothelium Table 1 HSP27 and HSP 70 expression by immunohistochemistry Percentage of reddish stained structures in total
Graduation
Explanation
0
Grade 0
No reaction
[0–29.99
Grade 1
Weak reaction
30–59.99 60–100
Grade 2 Grade 3
Moderate staining Intensive staining
Grade 4
Analyzed structures were not present in the section of the tissue sample (i.e., a central bronchus was not found in the sample)
Control group In all 29 cases, HE-stained sections showed no remarkable abnormal structures in tissues. In all cases, no HSP70-stained structures were observed in the tissues from hearts, lungs, or kidneys (Table 3). Of note, a strong positive staining for HSP70 was observed in one case of premortem heat exposure (Fig. 3a, b).
Table 3 Results for HSP70 staining Study group n = 120
Control group n = 29
Heart
Negative
Negative
Kidneys
Negative
Negative
Lungs
Negative
Negative
Table 2 Results for HSP27 staining Study group n = 120
Control group n = 29
Heart
HSP negative
4 % of endothelial cells stained positively
Kidneys
1 % of tubular cells stained positive
8 % of tubular cells stained positively 4 % of vascular endothelium cells stained positively
Lungs
27 % of respiratory epithelium cells stained positive
8 % of respiratory epithelium cells
19 % of vascular endothelium cells stained positive
29 % of vascular endothelium cells stained positively
123
36
Forensic Sci Med Pathol (2016) 12:33–39
Fig. 1 Positive immunostaining for HSP27. a Positive control. Subcutaneous tissue stained for HSP27 at 9200 magnification. The vascular wall is grade 2–3 positive. Vessel walls are shown (arrows). Normally erythrocytes are not stained but leukocytes may be stained,
b lung HSP27 at 9200 magnification, respiratory epithelium grade 1 positive, vascular wall grade 2 positive, c heart HSP27 at 9200 magnification, vascular wall grade 2 positive, d kidney HSP27 at 9200 magnification, tubular epithelium grade 1 positive
Discussion
In animal experiments on newborn rats and young mice, the inhibitory effects of hyperthermia on mechanisms involved in autoresuscitation from hypoxic apnea were demonstrated. The respiratory rate and frequencies of successful autoresuscitation were decreased, and the cardiac rate was increased [26, 39]. Immunohistochemical staining for HSP70 and HSP27 demonstrated that the hypothesis of hyperthermia being a risk factor for SIDS and playing a role in its pathogenesis was not confirmed. All HSP70 staining was negative and semiquantitative HSP27 staining in the SIDS group was not statistically different from that in the control group. This leads to the question whether hyperthermia really represents a risk factor in the pathogenesis of SIDS or whether immunohistochemical staining is not a suitable analysis method. One possible explanation for the negative results observed in this study might be that an elevated body core temperature in SIDS cases does not reach the threshold value for the expression of HSPs. Hyperthermic effects on cells at about 40 °C cause an increased incidence of protein denaturation and temperatures above 42.5 °C have direct cytotoxic effects [40, 41]. HSPs are rapidly expressed in cases of intense hyperthermia as we previously reported for
Hyperthermia enhances metabolism and leads to increased oxygen demand and respiratory rate. Hyperthermia in combination with hypoxia and hypercapnia might also lead to death in cases of prone position or with an infant being under a soft cover. Hyperthermia experiments showed that animals died after several hours of being positioned under a feather pillow [37]. A prone position can cause disturbances to physiological thermoregulation [14, 15] with an increased body surface temperature and increased cardiac and respiratory rates [12]. In a supine position, the body temperature is better regulated than in a prone position owing to undisturbed mobility [15]. In the prone position, the contact areas of the face and body with the underlying support are larger, which significantly impedes heat dissipation [9, 14, 17]. It was reported the risk of SIDS was increased by a factor of 6–10 in the prone position [18]. The inability of autoresuscitation with waking up and returning to a normal breathing rhythm is one cause of SIDS that has been discussed in cases of prolonged sleep apnea [5, 38]. The influence of a prone position and thermal stress on autoresuscitation in a hypoxic state was also reported [39].
123
Forensic Sci Med Pathol (2016) 12:33–39
37
Fig. 2 Immunostaining for HSP70. a Positive control. Subcutaneous tissue stained for HSP70 at 9200 magnification, b lung HSP70 at 9200 magnification was negative
Fig. 3 a Premortem heat exposure with short survival time (fire death). Kidney: glomeruli and tubular epithelium clearly positive for HSP70 staining at 9200 magnification. The staining in the glomerulus is spotty (positive staining of glomerular podocytes), b premortem
heat exposure with short survival time (fire death). Kidney: glomeruli and tubular epithelia positive for HSP70 staining at 9400 magnification
victims of fire death with very short terminal episodes [35] (Fig. 3a, b). HSPs are also expressed in cases of hypothermia; however, in cases of hypothermia, a long-lasting terminal episode of several hours is usually observed and death occurs at body core temperatures at around 25 °C. This is a deviation of 11–12 °C compared with the normal body core temperature. Temperature increases above 42.5 °C have direct cytotoxic effects dependent on temperature impact and temperature/time exposure. Furthermore, cells with continuous sublethal temperatures \42.5 °C may develop thermoresistance [41]. The longer and greater the initial temperature stress is, the longer the phase of thermoresistance lasts. Future prospective studies including the measurements of SIDS victim body core temperatures as soon as possible after death, followed by immunohistochemical stainings, might provide more information, and allow the determination of whether an elevated body core temperature correlates with the higher expression of HSPs. Another issue that might have a negative influence on the results could be the long storage period of organs in formaldehyde prior to this study. Because the number of
SIDS cases has, fortunately, decreased considerably over the years, our institute only examines an average of five SIDS cases per year; thus, the 120 cases in this study were collected from 1996 until 2013. The HSP proteins measured might have denatured and are thus no longer detectable by immunohistochemical staining. However, because there were no detectable differences between the oldest and the most recent tissue samples regarding immunohistochemical stainings for HSP70, this possibility seems unlikely. This was also confirmed by a study on the expression of HSP27 and HSP70 in cases of death due to fire [35]. Furthermore, previous studies reported that gene polymorphisms with a lack of HSP expression or expression of defective HSP were predisposing factors for SIDS [31].
Conclusions Significant premortem hyperthermia could not be confirmed by immunohistochemical examination of postmortem HSP expression in SIDS cases. This may be because the temperature level required to induce the
123
38
Forensic Sci Med Pathol (2016) 12:33–39
expression of HSPs was not reached or surpassed, or because the agonal periods (temperature time exposures) were not long enough. In contrast, in fire-related deaths, HSP27 and HSP70 are rapidly expressed after heat exposure at a higher grade. However, it should also be noted that SIDS does not represent a uniform disorder but rather a type of heterogeneous disease with various triggering and potentiating factors.
Key points 1. 2.
3. 4.
5.
6.
Hyperthermia has been considered a risk factor for SIDS. In cases of stress, including thermal stress, the expression of heat-shock proteins increases as a cellular stress response. In cases of lethal hypothermia as well as in fire victims, strong HSP expression was observed. The hypothesis of hyperthermia being a pathogenic factor in SIDS deaths was examined by immunohistochemical staining of HSP expression in lungs, hearts, and kidneys. Stainings for HSP70 were negative, and for HSP27 there was no difference compared with the control group. The hypothesis of hyperthermia as a pathogenetic factor for SIDS cannot be confirmed by the immunohistochemical visualization of HSP27 and HSP70.
References 1. Bajanowski T, Poets C. Der plo¨tzliche Sa¨uglingstod: Epidemiologie, a¨tiologie, pathophysiologie und Differenzialdiagnostik. Dtsch Arztebl. 2004;101:A-3185/B-2695/C-2567. 2. Byard R. Sudden death in the young. 3rd ed. Cambridge: University Press; 2010. 3. Doberentz E, Genneper L, Madea B. Plo¨tzlicher Sa¨uglingstod. Keine renale Hitzeschockproteinexpression. Rechtsmedizin. 2011;21:522–6. 4. Galland BC, Taylor BJ, Bolton DPG. Prone versus supine sleep position: the review of the physiological studies in SIDS research. J Paediatr Child Health. 2002;38:332–8. 5. Hunt CE, Brouillette RT. Sudden infant death syndrome: 1987 perspective. J Pediatr. 1987;110:669–78. 6. Hunt CE, Hauck FR. Sudden infant death syndrome. CMAJ. 2006;174:1861–9. 7. Sperhake JP. Pra¨vention des plo¨tzlichen Sa¨uglingstods (SIDS) in Hamburg 1995–2006. Eine populationsbasierte beobachtungspraxenstudie. Hamburg: Habilitationsschrift; 2008. 8. Vennemann MM, Bajanowski T, Brinkmann B, Jorch G, Sauerland C, Mitchell EA, GeSID Study Group. Sleep environment risk factors for sudden infant death syndrome: the German sudden infant death syndrome study. Pediatrics. 2009;123:1162–70.
123
9. Fleming PJ, Levine MR, Azaz Y, Wigfield R. The development of thermoregulation and interactions with the control of respiration in infants: possible relationship to sudden infant death. Acta Paediatr Scand Suppl. 1993;389:57–9. 10. Gilbert R, Rudd P, Berry P, Fleming PJ, Hall E, White D, et al. Combined effect of infection and heavy wrapping on the risk of sudden unexpected infant death. Arch Dis Child. 1992;67:171–7. 11. Kleemann WJ, Schlaud M, Poets CF, Rotha¨mel T, Tro¨ger HD. Hyperthermia in sudden infant death. Int J Legal Med. 1996;109:139–42. 12. Ammari A, Schulze KF, Ohira-Kist K, Kashyap S, Fifer WP, Myers MM, et al. Effects of body position on thermal, cardioresiratory and metabolic activity in low birth weight infants. Early Hum Dev. 2009;85:497–501. 13. Harper RM, Kinney HC, Fleming PJ, Thach BT. Sleep influences on homeostatic functions: implications for sudden infant death syndrome. Respir Physiol. 2000;119:123–32. 14. Nelson EAS, Taylor BJ, Weatherall IL. Sleeping position and infant bedding may predispose to hyperthermia and the sudden infant death syndrome. Lancet. 1989;1:199–202. 15. North RG, Petersen SA, Wailoo MP. Lower body temperature in sleeping supine infants. Arch Dis Child. 1995;72:340–2. 16. Petersen S, Anderson ES, Lodemore M, Rawson D, Wailoo MP. Sleeping position and rectal temperature. Arch Dis Child. 1991;66:976–9. 17. Ponsonby AL, Dwyer T, Gibbons LE, Cochrane JA, Jones ME, Mc Call MJ. Thermal environment and sudden infant death syndrome: case-control study. Br Med J. 1992;304:277–83. 18. Ponsonby AL, Dwyer T, Gibbons LE, Cochrane JA, Wang YG. Factors potentiating the risk of sudden infant death syndrome associated with the prone position. N Engl J Med. 1993;329:377–82. 19. Pfeifer K. Bedeutung der Rektaltemperaturmessung und der Umgebungstemperaturen beim plo¨tzlichen Kindstod. Dtsch Med Wochenschr. 1980;105:1066. 20. Pfeifer K, Pfeifer I, Adamczyk G, Adamczyk B. Probleme des plo¨tzlichen und unerwarteten Kindstodes aus der Sicht des Pathologen und Mikrobiologen. Dtsch Gesundheitswesen. 1977;32:317–32. 21. Agashe VR, Hartl FU. Roles of molecular chaperones in cytoplasmic protein folding. Semin Cell Dev Biol. 2000;11:15–25. 22. Beck FX, Neuhofer W, Mu¨ller E. Molecular chaperones in the kidney: distribution, putative roles, and regulation. Am J Physiol Renal Physiol. 2000;279:F203–15. 23. Flanagan SW, Ryan AJ, Gisolfi CV, Moseley PL. Tissue-specific HSP70 response in animals undergoing heat stress. Am J Physiol Regul. 1995;268:R28–32. 24. Hall DM, Xu L, Drake VJ, Oberley TD, Moseley PL, Kregel KC. Aging reduces adaptive capacity and stress protein expression in the liver after heat stress. J Appl Physiol. 2000;89:749–59. 25. Henle KJ, Leeper DB. Modification of the heat response and thermotolerance by cycloheximide, hydroxyurea and lucanthone in CHO cells. Radiat Res. 1982;90:339–47. 26. Kahraman L, Thach BT. Inhibitory effects of hyperthermia on mechanisms involved in autoresuscitation from hypoxic apnea in mice: a model for thermal stress causing SIDS. J Appl Physiol. 2004;97:669–74. 27. Katschinski DM. On heat and cells and proteins. News Physiol Sci. 2004;19:11–5. 28. Madden LA, Sandstro¨m MA, Lovell RJ, McNaughton L. Inducible heat shock protein 70 and its role in preconditioning and exercise. Amino Acids. 2008;34:511–6. 29. Marshall S, Rothschild MA, Bohnert M. Expression of heatshock protein 89 (Hsp70) in the respiratory tract and lungs of fire victims. Int J Legal Med. 2006;120(6):355–9.
Forensic Sci Med Pathol (2016) 12:33–39 30. Morimoto R, Tissieres A, Georgopoulos C. Heat shock proteins: structure, function and regulation. New York: Cold Spring Harbor Laboratory Press; 1994. 31. Rahim RA, Boyd PA, Ainslie Patrick WJ, Burdon RH. Human heat shock protein gene polymorphisms and sudden infant death syndrome. Arch Dis Child. 1996;75:451–2. 32. Ritossa F. A new puffing pattern induced by temperature shock and DNA in Drosophila. Experientia. 1962;18:571–3. 33. Ryan AJ, Gisolfi CV, Moseley PL. Synthesis of 70 K stress protein by human leukocytes: effect of exercise in the heat. J Appl Physiol. 1991;70:466–71. 34. Preuss J, Dettmeyer R, Poster S, Lignitz E, Madea M. The expression of heat shock protein 70 in kidneys in cases of death due to hyperthermia. Forensic Sci Int. 2008;176:248–52. 35. Doberentz E, Genepper L, Bo¨ker D, Lignitz E, Madea B. Expression of heat shock protein (hsp) 27 and 70 in various organ systems in cases of death due to fire. Int J Legel Med. 2014;128:967–78. 36. Krous HF, Beckwith JB, Byard RW, Rognum TO, Bajanowski T, Corey T, et al. Sudden infant death syndrome and unclassified
39
37.
38.
39.
40.
41.
sudden infant death: a definitional and diagnostic approach. Pediatrics. 2004;114:234–8. Tausch D, Mo¨ller M. Tierexperimentelle Untersuchungen u¨ber Sauerstoffmangelzusta¨nde im Hinblick auf den‘‘plo¨tzlichen Kindstod’’. Beitr Gerichtl Med. 1973;31:130–4. Poets LF, Meny RG, Chobanian MR, Bouofiglo RE. Gasping and other respirators patterns during sudden infant deaths. Pediatr Res. 1999;45:350–5. Sederevich C, Fewell JE. Influence of core temperature on the time to last gasp and autoresuscitation from primary apnea during exposure to hypoxia in normal rat pups. J Appl Physiol. 1999;87:1346–53. Becker VT. Einfluss der Temperatur auf die Lungenfunktionsparameter im Rahmen der isolierten hyperthermen Lungenperfusion. Dissertation. Jena. 2004. Issels RD, Hegewisch-Becker S. Chemotherapie in Kombination mit Hyperthermie. In: Schmoll HJ, Ho¨ffken K, Possinger K, editors. Kompendium internistische Onkologie. Standards in Diagnostik und Therapie. 4th ed. Heidelberg: Springer; 2006. p. 1091–108.
123
ORIGINAL ARTICLE
Sudden infant death syndrome: no significant expression of heat-shock proteins (HSP27, HSP70) Elke Doberentz1 • Sarah Fu¨hring1 • Burkhard Madea1
Accepted: 16 November 2015 / Published online: 12 December 2015 Ó Springer Science+Business Media New York 2015
Abstract Purpose In industrialized countries, sudden infant death is the most common cause of death in young children. Although prone sleeping position is a well-known risk factor, hyperthermia might also be important. Pathognomonic findings of premortem hyperthermia do not exist. During stress, including thermal effects, heat-shock protein (HSP) expression increases. This study investigated hyperthermia as a contributing or pathogenic factor for sudden infant death syndrome (SIDS). Methods Immunohistochemical staining for HSP27 and HSP70 in the kidney, heart, and lung from 120 SIDS cases was examined. Results HSP70 immunostaining was negative in kidney, heart, and lung tissues in all cases and in tissues from the control group. HSP27 staining was positive in the kidney from one case, and was positive in the lungs (respiratory epithelia in 27 % of cases; vascular endothelia in 19 % of cases) and was negative in the heart. In the control group HSP27 was positive in 8 % of renal tubular tissues and in 29 % of renal vascular endothelia. Staining for HSP27 in lung tissues was positive in respiratory epithelia in 8 % of cases and for vascular endothelia in 29 %, whereas tissues from the heart were positive in only 4 %. Conclusion The hypothesis of hyperthermia being a pathogenic factor for SIDS was not supported by immunohistochemical visualization of HSP70 or HSP27.
& Burkhard Madea [email protected] 1
Institute of Forensic Medicine, University of Bonn, Stiftsplatz 12, 53111 Bonn, Germany
Keywords SIDS Hyperthermia Body temperature Stress Immunohistochemistry Heat-shock proteins
Introduction Sudden infant death syndrome (SIDS) involves the sudden and unexpected death of apparently healthy infants. Despite extensive postmortem examinations of SIDS infants, the cause of death is poorly understood. In industrialized countries, sudden infant death is the most common cause of death among children during their first years of life. Several risk factors for SIDS have been identified, and the most well known is prone sleeping position [1–8]. Besides prone sleeping position, hyperthermia is also considered a possible risk factor. Thermal stress may increase the risk of SIDS, for example from high ambient temperatures, warm clothing or bedding, and increased body temperature during infectious diseases [2, 9–11]. Interestingly, sleeping position itself may predispose to hyperthermia [11–18]. In a study by Pfeifer on 107 of 142 postmortem SIDS babies, rectal body temperatures over 40 °C were measured [19, 20]. According to Kleemann et al. [11] the role of hyperthermia in sudden infant death has been discussed for many years. As a result of structured interviews with parents of SIDS victims (n = 140), signs of profuse sweating (moist head and hair and damp clothing or bedding) were noted at the death scenes in 35.7 % of cases compared to controls in the same geographical region. Sweat on the head and sweaty clothing has a significant association with the risk of SIDS. According to these studies, the pathophysiological basis for hyperthermia remains unclear. It might result from infections, over insulation, excessive clothing in high
123
34
environmental temperatures, covering of an infant’s head or immature central thermoregulatory centers. A case–control study on sleep environment risk factors for SIDS revealed that duvets double the risk for SIDS compared with infants using a sleeping bag only, or a very light cotton blanket [8]. Another case–control study confirmed that SIDS cases were characterized by excess thermal insulation for the given room temperatures compared with matched controls [17]. Overheating was independent to prone sleeping position associated with an increased risk for sudden infant death. In summary, there is evidence from a number of scene findings as well as from case–control studies that some infants experience thermal stress and that infant hyperthermia may have a role in the etiology of SIDS. However, according to Byard [2] it is likely that hyperthermia might be a risk modifier in infants who sleep in a prone position. This conclusion is based on the findings of Ponsonby et al. [17] who showed that SIDS cases are significantly more likely to be overdressed for the room temperature at the death scene and who were sleeping in the prone position compared with control infants. In cases of stress, including thermal effects, the expression of heat-shock proteins (HSP) may increase as a response to cellular stress. HSP are rapidly expressed with both endogenous and exogenous cellular stress. They support protein folding of newly synthesized proteins and are important for the stabilization and degradation of incorrectly folded or damaged proteins. They are divided into subgroups according to their molecular weights and can be stained for immunohistochemically [21–33]. HSPs are also expressed in renal tissue in cases of lethal hypothermia [34]. In a study group of 100 cases of death due to hypothermia, renal glomerular expression of HSP70 was present in 80 % of cases, and renal tubular expression was found in 89 % of cases compared with the control group (n = 50) with short and long agonal periods where no renal glomerular expression was found in 84 % of cases and no renal tubular expression was observed in 66 % of all cases [34]. HSPs are also rapidly expressed in various organ systems in cases of death due to fire [35]. HSP27 has a molecular weight of 27 kDa and belongs to a group of small HSP. Small HSP are present in the cytosol of most cells, even in the absence of stress factors; under heat-shock conditions their expression is increased dramatically. Their main purpose is the stabilization of microfilaments. HSP70 has a molecular weight of 70–78 kDa. At least eight subgroups have been identified in humans. HSP70 is expressed in the cytosol, nucleus, mitochondria, and endoplasmic reticulum of the cell and accounts for 1–2 % of the total cellular proteins. Their main purpose is protein folding, protein translocation, and cytoprotection.
123
Forensic Sci Med Pathol (2016) 12:33–39
The objective of the current study was to determine whether increased HSP27 or HSP70 expression is associated with premortem hyperthermia in SIDS. Immunohistochemical staining for HSP70 and HSP27 was performed on the hearts, lungs, and especially the kidneys, because they have a rapid HSP reaction after thermal stress [35].
Materials and methods The study group consisted of 120 SIDS cases (46 female and 74 male individuals). The diagnoses of SIDS as a cause of death were based upon the criteria of the San Diego Consensus Definition from 2004 [36]. Most cases belonged to category Ib. The minimum age was 7 days and the maximum was 344 days (mean age 125 days). The control group consisted of 29 cases of death in children due to unnatural causes (13 female and 16 male individuals) with a minimum age of 4 days and a maximum age of 354 days. During each autopsy, organ samples were obtained and stored in formaldehyde solution until further examination. For this study, the tissues were first dehydrated using an increasing sequence of alcohol solutions. Subsequently they were embedded in paraffin. The tissue blocks were then cut into 2–4 lm-thin slices and transferred to microscope slides. After drying in an incubator, the slides were dewaxed and rehydrated. Immunohistochemical staining for HSP70 and HSP27 were carried out according to the labeled streptavidin–biotin method with specific primary antibodies (Novo CastraÒ HSP27 mouse monoclonal antibody; Novo CastraÒ HSP70 mouse monoclonal antibody), a biotinylated secondary antibody and a labeled streptavidin complex. Staining of the specimens was completed with a chromogen substrate (Dako AEC Substrate Chromogen Code K3464) and hemalum for staining of the nuclei. Positive and negative controls were used in each staining process to verify the findings and to confirm antibody specificity. For examination of the stained samples, 30 visual fields of each specimen were semiquantitatively evaluated at 4009 magnification. Heart, lungs, and especially kidneys were chosen for investigation, because kidneys have a rapid HSP expression after thermal stress. Evaluation of the heart with epicardium, myocytes disci intercalares, perinuclear spaces, vessels, and fibrocytes; lungs with peripheral and central bronchial tubes, vessels (endothelia, lumina), interalveolar septa, pleura, peribronchial glands, peribronchial connective tissues, ciliated epithelia as well as the central respiratory tracts with bronchial tubes, ciliated epithelia, peribronchial glands, vessels (endothelia, lumina), cartilages, glandular tissues, and lymphatic tissues; and the kidneys with renal tube glomeruli, vessels, capsules, and connective tissues were used for immunohistochemical analysis, which was graded
Forensic Sci Med Pathol (2016) 12:33–39
semiquantitatively using the four-degree scale reported by Preuss et al. [34]. The numbers of positive red stained structures/cells observed in each visual field were estimated as a percentage of all investigated structures/cells. For each analyzed slice and structure, a mean value from all 30 visual fields was calculated and graduated according to the pattern in Table 1.
Results Tables 2 and 3 show the immunohistochemical analysis results for HSP70 and HSP27 in the cellular structures of each organ. Representative images of the immunohistochemical analysis for HSP27 and HSP70 stainings are shown in Figs. 1 and 2.
35
cells were stained positive for HSP27, respectively (Fig. 1a–d; Table 2). Control group In all 29 cases, the HE-stained sections showed no remarkable abnormal structures in tissues. In 4 % of cases, heart endothelium cells were stained positive for HSP27. In 8 and 4 % of cases, kidney tubular cells and vascular endothelium cells were stained positive for HSP27, respectively. In 8 and 29 % of cases, lung respiratory epithelium cells and vascular endothelium cells were stained positive for HSP27, respectively (Table 2).
HSP70 immunohistochemical analysis
HSP27 immunohistochemical analysis
Study group
Study group
In all 120 cases, HE-stained sections showed no remarkable abnormal structures in tissues. In all cases, no HSP70 stained structures were observed in the tissues from hearts, lungs, or kidneys (Fig. 2a, b; Table 3).
In all 120 cases, the hematoxylin and eosin (HE)-stained sections showed no remarkable abnormal structures in tissues. No stained structures were observed in tissue samples from hearts. In 1 % of cases, kidney tubular cells were stained positive for HSP27. In 27 and 19 % of cases, lung respiratory epithelium cells and vascular endothelium Table 1 HSP27 and HSP 70 expression by immunohistochemistry Percentage of reddish stained structures in total
Graduation
Explanation
0
Grade 0
No reaction
[0–29.99
Grade 1
Weak reaction
30–59.99 60–100
Grade 2 Grade 3
Moderate staining Intensive staining
Grade 4
Analyzed structures were not present in the section of the tissue sample (i.e., a central bronchus was not found in the sample)
Control group In all 29 cases, HE-stained sections showed no remarkable abnormal structures in tissues. In all cases, no HSP70-stained structures were observed in the tissues from hearts, lungs, or kidneys (Table 3). Of note, a strong positive staining for HSP70 was observed in one case of premortem heat exposure (Fig. 3a, b).
Table 3 Results for HSP70 staining Study group n = 120
Control group n = 29
Heart
Negative
Negative
Kidneys
Negative
Negative
Lungs
Negative
Negative
Table 2 Results for HSP27 staining Study group n = 120
Control group n = 29
Heart
HSP negative
4 % of endothelial cells stained positively
Kidneys
1 % of tubular cells stained positive
8 % of tubular cells stained positively 4 % of vascular endothelium cells stained positively
Lungs
27 % of respiratory epithelium cells stained positive
8 % of respiratory epithelium cells
19 % of vascular endothelium cells stained positive
29 % of vascular endothelium cells stained positively
123
36
Forensic Sci Med Pathol (2016) 12:33–39
Fig. 1 Positive immunostaining for HSP27. a Positive control. Subcutaneous tissue stained for HSP27 at 9200 magnification. The vascular wall is grade 2–3 positive. Vessel walls are shown (arrows). Normally erythrocytes are not stained but leukocytes may be stained,
b lung HSP27 at 9200 magnification, respiratory epithelium grade 1 positive, vascular wall grade 2 positive, c heart HSP27 at 9200 magnification, vascular wall grade 2 positive, d kidney HSP27 at 9200 magnification, tubular epithelium grade 1 positive
Discussion
In animal experiments on newborn rats and young mice, the inhibitory effects of hyperthermia on mechanisms involved in autoresuscitation from hypoxic apnea were demonstrated. The respiratory rate and frequencies of successful autoresuscitation were decreased, and the cardiac rate was increased [26, 39]. Immunohistochemical staining for HSP70 and HSP27 demonstrated that the hypothesis of hyperthermia being a risk factor for SIDS and playing a role in its pathogenesis was not confirmed. All HSP70 staining was negative and semiquantitative HSP27 staining in the SIDS group was not statistically different from that in the control group. This leads to the question whether hyperthermia really represents a risk factor in the pathogenesis of SIDS or whether immunohistochemical staining is not a suitable analysis method. One possible explanation for the negative results observed in this study might be that an elevated body core temperature in SIDS cases does not reach the threshold value for the expression of HSPs. Hyperthermic effects on cells at about 40 °C cause an increased incidence of protein denaturation and temperatures above 42.5 °C have direct cytotoxic effects [40, 41]. HSPs are rapidly expressed in cases of intense hyperthermia as we previously reported for
Hyperthermia enhances metabolism and leads to increased oxygen demand and respiratory rate. Hyperthermia in combination with hypoxia and hypercapnia might also lead to death in cases of prone position or with an infant being under a soft cover. Hyperthermia experiments showed that animals died after several hours of being positioned under a feather pillow [37]. A prone position can cause disturbances to physiological thermoregulation [14, 15] with an increased body surface temperature and increased cardiac and respiratory rates [12]. In a supine position, the body temperature is better regulated than in a prone position owing to undisturbed mobility [15]. In the prone position, the contact areas of the face and body with the underlying support are larger, which significantly impedes heat dissipation [9, 14, 17]. It was reported the risk of SIDS was increased by a factor of 6–10 in the prone position [18]. The inability of autoresuscitation with waking up and returning to a normal breathing rhythm is one cause of SIDS that has been discussed in cases of prolonged sleep apnea [5, 38]. The influence of a prone position and thermal stress on autoresuscitation in a hypoxic state was also reported [39].
123
Forensic Sci Med Pathol (2016) 12:33–39
37
Fig. 2 Immunostaining for HSP70. a Positive control. Subcutaneous tissue stained for HSP70 at 9200 magnification, b lung HSP70 at 9200 magnification was negative
Fig. 3 a Premortem heat exposure with short survival time (fire death). Kidney: glomeruli and tubular epithelium clearly positive for HSP70 staining at 9200 magnification. The staining in the glomerulus is spotty (positive staining of glomerular podocytes), b premortem
heat exposure with short survival time (fire death). Kidney: glomeruli and tubular epithelia positive for HSP70 staining at 9400 magnification
victims of fire death with very short terminal episodes [35] (Fig. 3a, b). HSPs are also expressed in cases of hypothermia; however, in cases of hypothermia, a long-lasting terminal episode of several hours is usually observed and death occurs at body core temperatures at around 25 °C. This is a deviation of 11–12 °C compared with the normal body core temperature. Temperature increases above 42.5 °C have direct cytotoxic effects dependent on temperature impact and temperature/time exposure. Furthermore, cells with continuous sublethal temperatures \42.5 °C may develop thermoresistance [41]. The longer and greater the initial temperature stress is, the longer the phase of thermoresistance lasts. Future prospective studies including the measurements of SIDS victim body core temperatures as soon as possible after death, followed by immunohistochemical stainings, might provide more information, and allow the determination of whether an elevated body core temperature correlates with the higher expression of HSPs. Another issue that might have a negative influence on the results could be the long storage period of organs in formaldehyde prior to this study. Because the number of
SIDS cases has, fortunately, decreased considerably over the years, our institute only examines an average of five SIDS cases per year; thus, the 120 cases in this study were collected from 1996 until 2013. The HSP proteins measured might have denatured and are thus no longer detectable by immunohistochemical staining. However, because there were no detectable differences between the oldest and the most recent tissue samples regarding immunohistochemical stainings for HSP70, this possibility seems unlikely. This was also confirmed by a study on the expression of HSP27 and HSP70 in cases of death due to fire [35]. Furthermore, previous studies reported that gene polymorphisms with a lack of HSP expression or expression of defective HSP were predisposing factors for SIDS [31].
Conclusions Significant premortem hyperthermia could not be confirmed by immunohistochemical examination of postmortem HSP expression in SIDS cases. This may be because the temperature level required to induce the
123
38
Forensic Sci Med Pathol (2016) 12:33–39
expression of HSPs was not reached or surpassed, or because the agonal periods (temperature time exposures) were not long enough. In contrast, in fire-related deaths, HSP27 and HSP70 are rapidly expressed after heat exposure at a higher grade. However, it should also be noted that SIDS does not represent a uniform disorder but rather a type of heterogeneous disease with various triggering and potentiating factors.
Key points 1. 2.
3. 4.
5.
6.
Hyperthermia has been considered a risk factor for SIDS. In cases of stress, including thermal stress, the expression of heat-shock proteins increases as a cellular stress response. In cases of lethal hypothermia as well as in fire victims, strong HSP expression was observed. The hypothesis of hyperthermia being a pathogenic factor in SIDS deaths was examined by immunohistochemical staining of HSP expression in lungs, hearts, and kidneys. Stainings for HSP70 were negative, and for HSP27 there was no difference compared with the control group. The hypothesis of hyperthermia as a pathogenetic factor for SIDS cannot be confirmed by the immunohistochemical visualization of HSP27 and HSP70.
References 1. Bajanowski T, Poets C. Der plo¨tzliche Sa¨uglingstod: Epidemiologie, a¨tiologie, pathophysiologie und Differenzialdiagnostik. Dtsch Arztebl. 2004;101:A-3185/B-2695/C-2567. 2. Byard R. Sudden death in the young. 3rd ed. Cambridge: University Press; 2010. 3. Doberentz E, Genneper L, Madea B. Plo¨tzlicher Sa¨uglingstod. Keine renale Hitzeschockproteinexpression. Rechtsmedizin. 2011;21:522–6. 4. Galland BC, Taylor BJ, Bolton DPG. Prone versus supine sleep position: the review of the physiological studies in SIDS research. J Paediatr Child Health. 2002;38:332–8. 5. Hunt CE, Brouillette RT. Sudden infant death syndrome: 1987 perspective. J Pediatr. 1987;110:669–78. 6. Hunt CE, Hauck FR. Sudden infant death syndrome. CMAJ. 2006;174:1861–9. 7. Sperhake JP. Pra¨vention des plo¨tzlichen Sa¨uglingstods (SIDS) in Hamburg 1995–2006. Eine populationsbasierte beobachtungspraxenstudie. Hamburg: Habilitationsschrift; 2008. 8. Vennemann MM, Bajanowski T, Brinkmann B, Jorch G, Sauerland C, Mitchell EA, GeSID Study Group. Sleep environment risk factors for sudden infant death syndrome: the German sudden infant death syndrome study. Pediatrics. 2009;123:1162–70.
123
9. Fleming PJ, Levine MR, Azaz Y, Wigfield R. The development of thermoregulation and interactions with the control of respiration in infants: possible relationship to sudden infant death. Acta Paediatr Scand Suppl. 1993;389:57–9. 10. Gilbert R, Rudd P, Berry P, Fleming PJ, Hall E, White D, et al. Combined effect of infection and heavy wrapping on the risk of sudden unexpected infant death. Arch Dis Child. 1992;67:171–7. 11. Kleemann WJ, Schlaud M, Poets CF, Rotha¨mel T, Tro¨ger HD. Hyperthermia in sudden infant death. Int J Legal Med. 1996;109:139–42. 12. Ammari A, Schulze KF, Ohira-Kist K, Kashyap S, Fifer WP, Myers MM, et al. Effects of body position on thermal, cardioresiratory and metabolic activity in low birth weight infants. Early Hum Dev. 2009;85:497–501. 13. Harper RM, Kinney HC, Fleming PJ, Thach BT. Sleep influences on homeostatic functions: implications for sudden infant death syndrome. Respir Physiol. 2000;119:123–32. 14. Nelson EAS, Taylor BJ, Weatherall IL. Sleeping position and infant bedding may predispose to hyperthermia and the sudden infant death syndrome. Lancet. 1989;1:199–202. 15. North RG, Petersen SA, Wailoo MP. Lower body temperature in sleeping supine infants. Arch Dis Child. 1995;72:340–2. 16. Petersen S, Anderson ES, Lodemore M, Rawson D, Wailoo MP. Sleeping position and rectal temperature. Arch Dis Child. 1991;66:976–9. 17. Ponsonby AL, Dwyer T, Gibbons LE, Cochrane JA, Jones ME, Mc Call MJ. Thermal environment and sudden infant death syndrome: case-control study. Br Med J. 1992;304:277–83. 18. Ponsonby AL, Dwyer T, Gibbons LE, Cochrane JA, Wang YG. Factors potentiating the risk of sudden infant death syndrome associated with the prone position. N Engl J Med. 1993;329:377–82. 19. Pfeifer K. Bedeutung der Rektaltemperaturmessung und der Umgebungstemperaturen beim plo¨tzlichen Kindstod. Dtsch Med Wochenschr. 1980;105:1066. 20. Pfeifer K, Pfeifer I, Adamczyk G, Adamczyk B. Probleme des plo¨tzlichen und unerwarteten Kindstodes aus der Sicht des Pathologen und Mikrobiologen. Dtsch Gesundheitswesen. 1977;32:317–32. 21. Agashe VR, Hartl FU. Roles of molecular chaperones in cytoplasmic protein folding. Semin Cell Dev Biol. 2000;11:15–25. 22. Beck FX, Neuhofer W, Mu¨ller E. Molecular chaperones in the kidney: distribution, putative roles, and regulation. Am J Physiol Renal Physiol. 2000;279:F203–15. 23. Flanagan SW, Ryan AJ, Gisolfi CV, Moseley PL. Tissue-specific HSP70 response in animals undergoing heat stress. Am J Physiol Regul. 1995;268:R28–32. 24. Hall DM, Xu L, Drake VJ, Oberley TD, Moseley PL, Kregel KC. Aging reduces adaptive capacity and stress protein expression in the liver after heat stress. J Appl Physiol. 2000;89:749–59. 25. Henle KJ, Leeper DB. Modification of the heat response and thermotolerance by cycloheximide, hydroxyurea and lucanthone in CHO cells. Radiat Res. 1982;90:339–47. 26. Kahraman L, Thach BT. Inhibitory effects of hyperthermia on mechanisms involved in autoresuscitation from hypoxic apnea in mice: a model for thermal stress causing SIDS. J Appl Physiol. 2004;97:669–74. 27. Katschinski DM. On heat and cells and proteins. News Physiol Sci. 2004;19:11–5. 28. Madden LA, Sandstro¨m MA, Lovell RJ, McNaughton L. Inducible heat shock protein 70 and its role in preconditioning and exercise. Amino Acids. 2008;34:511–6. 29. Marshall S, Rothschild MA, Bohnert M. Expression of heatshock protein 89 (Hsp70) in the respiratory tract and lungs of fire victims. Int J Legal Med. 2006;120(6):355–9.
Forensic Sci Med Pathol (2016) 12:33–39 30. Morimoto R, Tissieres A, Georgopoulos C. Heat shock proteins: structure, function and regulation. New York: Cold Spring Harbor Laboratory Press; 1994. 31. Rahim RA, Boyd PA, Ainslie Patrick WJ, Burdon RH. Human heat shock protein gene polymorphisms and sudden infant death syndrome. Arch Dis Child. 1996;75:451–2. 32. Ritossa F. A new puffing pattern induced by temperature shock and DNA in Drosophila. Experientia. 1962;18:571–3. 33. Ryan AJ, Gisolfi CV, Moseley PL. Synthesis of 70 K stress protein by human leukocytes: effect of exercise in the heat. J Appl Physiol. 1991;70:466–71. 34. Preuss J, Dettmeyer R, Poster S, Lignitz E, Madea M. The expression of heat shock protein 70 in kidneys in cases of death due to hyperthermia. Forensic Sci Int. 2008;176:248–52. 35. Doberentz E, Genepper L, Bo¨ker D, Lignitz E, Madea B. Expression of heat shock protein (hsp) 27 and 70 in various organ systems in cases of death due to fire. Int J Legel Med. 2014;128:967–78. 36. Krous HF, Beckwith JB, Byard RW, Rognum TO, Bajanowski T, Corey T, et al. Sudden infant death syndrome and unclassified
39
37.
38.
39.
40.
41.
sudden infant death: a definitional and diagnostic approach. Pediatrics. 2004;114:234–8. Tausch D, Mo¨ller M. Tierexperimentelle Untersuchungen u¨ber Sauerstoffmangelzusta¨nde im Hinblick auf den‘‘plo¨tzlichen Kindstod’’. Beitr Gerichtl Med. 1973;31:130–4. Poets LF, Meny RG, Chobanian MR, Bouofiglo RE. Gasping and other respirators patterns during sudden infant deaths. Pediatr Res. 1999;45:350–5. Sederevich C, Fewell JE. Influence of core temperature on the time to last gasp and autoresuscitation from primary apnea during exposure to hypoxia in normal rat pups. J Appl Physiol. 1999;87:1346–53. Becker VT. Einfluss der Temperatur auf die Lungenfunktionsparameter im Rahmen der isolierten hyperthermen Lungenperfusion. Dissertation. Jena. 2004. Issels RD, Hegewisch-Becker S. Chemotherapie in Kombination mit Hyperthermie. In: Schmoll HJ, Ho¨ffken K, Possinger K, editors. Kompendium internistische Onkologie. Standards in Diagnostik und Therapie. 4th ed. Heidelberg: Springer; 2006. p. 1091–108.
123
