- Email: [email protected]
Incidence and predictive factors of isolated neonatal penile glanular torsion
Journal of Pediatric Urology (2007) 3, 495e499
Incidence and predictive factors of isolated neonatal penile glanular torsion Sarkis Pierrot E.*,1, Sadasivam Muthurajan1,2 Pediatric Surgery Section, Saad Specialist Hospital, Hammoud Street, P.O. Box: 30353, Al-Khobar 31952, Saudi Arabia Received 23 January 2007; accepted 7 March 2007 Available online 5 July 2007
KEYWORDS Penile malformations; Neonatal penile torsion; Median raphe deviation; Penile glanular torsion
Abstract Purpose: To determine the incidence of isolated neonatal penile glanular torsion, describe the basic characteristics, and explore the relationship between foreskin and glans torsion. Methodology: A prospective survey was conducted of all male newborns admitted to nursery after delivery, or neonates less than 3 months presenting for circumcision. Cases with associated genital malformations were excluded. Results: The incidence of isolated neonatal penile torsion was 27% (95% CI: 22.2%e31.84%), to the left in 99% of cases. In 3.5% of cases, the penis had an angle <10 , and 9.5% >20 . Using Spearman’s correlational coefficient, deviation of penile raphe from the midline at the foreskin tip had a better correlation with glans torsion than deviation of raphe at the coronal sulcus (0.727 vs 0.570; both significant at p < 0.01). Conclusion: Isolated neonatal penile torsion is more common than reported. The median raphe of the penis may be normal and mask unexpected glans torsion. Median raphe torsion at foreskin tip can be used as a predictor for glans torsion. Clinical significance and relation to adult penile torsion are beyond the scope of the study. ª 2007 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.
Introduction Penile torsion is a two- or three-dimensional rotation of the penis at the level of the shaft and/or the glans [1,2]. Pediatric surgeons frequently encounter penile torsion at the neonatal stage in countries where circumcision is routinely performed after delivery. The condition can be diagnosed * Corresponding author. Tel.: þ966 501617127/þ966 38014810. E-mail addresses: [email protected], psarkis@saad. com.sa (Sarkis P.E.), [email protected] (Sadasivam M.). 1 Tel.:þ966 3882666; fax: þ96638823334. 2 Tel.: þ966 38014807.
following delivery, or later during childhood. Neonatal penile torsion can be isolated or associated with male genital malformations [3], and the etiology is different from the adult type [4e7]. There are few publications on isolated neonatal penile torsion, and the clinical characteristics are not well known. The real incidence is difficult to evaluate for many reasons. First, isolated torsion should be separated from secondary cases due to associated penile anomalies like hypospadias [3]. Second, the deviation of the penile median raphe from the midline may or may not reflect the real angle of glanular torsion. Penile torsion should be assessed by measuring the angle of glanular torsion not foreskin alone.
1477-5131/$30 ª 2007 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company. doi:10.1016/j.jpurol.2007.03.002
496
Sarkis P. E., S. Muthurajan
We encountered glanular penile torsion more frequently than is reported [8]. The aims of this study were to determine the following. 1. The incidence of isolated neonatal penile glanular torsion, and its descriptive characteristics, in a country (Saudi Arabia) where circumcision is almost systematic at the neonatal stage mainly for ritual or religious reasons. 2. The correlation between glans torsion and foreskin torsion. 3. The possibility of using median raphe deviation or torsion as a predictor of glans torsion.
Patient and methods Between July and December 2005, a prospective survey was conducted at the Saad Specialist Hospital, AL-Khobar, KSA, of all newborn candidates for circumcision (Fig. 1).
Exclusion criteria 1. Newborn transferred to neonatal intensive care unit or with major medical pathology. 2. Refusal of neonatal circumcision by parents. 3. Insufficient data. 4. All secondary cases: defined as associated major congenital anomaly of genital organs, or coexistence of hypospadias, chordee or lateral curvature of penis.
Sampling and recruitment This was a hospital-based sample. A consecutive sampling technique was chosen because of absence of any seasonal factor or time-dependent trends, or any source of systematic error. Institutional review board and ethical committee approval was obtained for the study.
Strategies to enhance precision Inclusion criteria 1. Every male newborn admitted to nursery after delivery. 2. Or neonate less than 3 months of age, presenting to pediatric surgery outpatient clinic for circumcision. 3. Newborns with unilateral undescended testis, unilateral or bilateral hydrocele were not excluded.
Candidates for circumcision: 1- All male newborn delivered in SSH 2- OPD males <3 months old Between 15 July and 30 November 2005 (470 male/859 live births)
1st step exclusion criteria (83 boys): 1- Discharged against medical advice (24) or early discharge post delivery (18) 2- Transfer to NICU or any major medical contraindication (26) 3- Parents refuse circumcision (15)
All circumcisions and measurement were done by two surgeons personally. To ensure internal validity, and consistency between observers, the methodology of measurement has been standardized and recorded in writing. Sessions for practice of the technique were performed in the presence of both authors for the first 10 cases, and repeated every first week of each month.
Outcome of interest (endpoint) Isolated congenital neonatal penile torsion was measured by the degree of torsion at the level of the glans, which was compared to the angle of deviation of the median raphe at two different levels: coronal sulcus and foreskin tip (Figs. 2 and 3). For the purpose of this study, angles are grouped by intervals of 10 . To measure the angles with precision and
Eligible for the study (387 boys)
2nd step exclusion criteria (17 boys): 1- No or insufficient data (4) 2- Major congenital anomalies: ambiguous genitalia (1)… 3- Coronal or severe hypospadias (1) 4- Chordee (1) 5- Lateral curvature minor or severe (4) 6- Webbed penis (5) 7- No raphe apparent (1)
Final study population = 370 boys 1- Normal = 270 2- Isolated neonatal penile glanular torsion= 100
Figure 1 Flowchart of study population for isolated neonatal penile glanular torsion with inclusion and exclusion criteria. SSH, Saad Specialist Hospital; OPD, outpatient department; NICU, neonatal intensive care unit.
Level of measurement of median raphe deviation at foreskin tip
Level of measurement of median raphe deviation at coronal sulcus
Figure 2 Ventral median raphe highlighted to improve visualization and accuracy of measurement, and indication of the three levels of measurement (coronal sulcus, foreskin tip, and glans).
Neonatal penile glanular torsion
497 Table 1 Distribution of angle of median raphe deviation and glanular torsion (n Z 370)
Angle of torsion of meatus indicating the torsion of glans
Level of measurement of median raphe deviation at foreskin tip Level of measurement of median raphe deviation at coronal sulcus
Figure 3 Ventral median raphe highlighted to improve visualization and accuracy of measurement, and indication of the three levels of measurement (coronal sulcus, foreskin tip, and glans).
accuracy, the following technical adjustments were undertaken. 1. A small regular protractor was used, with minor modification for a better adjustment around the penile shaft. 2. Torsion angle of the foreskin: Before injection of penile block analgesia, the midline ventral penile raphe was highlighted to improve its visibility. The penis was put in a straight upright position by digital pressure on the dorsal root of the penis avoiding torsion. The angle of median raphe deviation from the midline at the level of the coronal sulcus and the tip was measured. 3. Torsion angle of the glans: The foreskin was split dorsally at the midline, and adhesiolysis of the glans was completed, as a first step for circumcision. Before any manipulation, the degree of torsion of the glans was measured at the level of the urethral meatus with a sterile protractor. For a descriptive study with 95% CI of 10%, and based on the possible incidence of neonatal penile torsion calculated from a pilot unpublished internal study, a sample of 330 male newborns was needed; this was achieved in 6 months. Statistical analysis was done using SPSS 14. Spearman nonparametric correlation coefficient between continuous or ordinal variables not normally distributed was calculated. To quantify the level of association and prediction of torsion, logistic regression analysis between the dependent ‘glans torsion’ and the independent angle of median raphe torsion at foreskin tip and coronal sulcus was done, and positive predictive values (compared by Chi-square test) were calculated. Significance level was fixed to two-tailed test with p < 0.05.
Results Data were collected on 387 male infant candidates for circumcision, of whom 105 had neonatal penile glanular torsion with an incidence of 27.4% (95% CI: 23%e31.8%). After exclusion of 17 files because of coexistence of genital
Angle ( )
Deviation of median raphe at coronal sulcus
Deviation of median raphe at foreskin tip
Torsion of glans
0 1e10 11e20 21e30 31e40 41e50 51e60 61e70 71e80 81e90
264 18 24 28 8 14 10 1 2 1
242 17 35 30 6 19 12 4 0 5
270 13 52 23 3 8 1 0 0 0
pathology (Fig. 1), the incidence of isolated neonatal penile torsion based on glans torsion was 27.02% (95% CI: 22.2%e 31.84%). Torsion was to the left in 99% of cases, in a counterclockwise direction, and in two cases to the right (2/370; <1%). Another case of right penile torsion was associated with right curvature and excluded from the study. Of the 370 included cases, 25 infants were non-Saudis. In the group of Saudi male newborns, penile glanular torsion was present in 82 cases (24%), with 10 out of 82 at less than 10 . The distribution of torsion at the three measured levels is shown for the whole population (Table 1) and for cases without glanular torsion (Table 2). No glans torsion exceeded 60 . There was special interest in the relationship between deviation of the median raphe and glans torsion. Even when there was no deviation from the midline, some glans showed unexpected torsion: in 4%e12%, depending on whether the principal indicator was fixed as the angle of deviation of median raphe at the coronal sulcus or foreskin tip (Tables 3A and 4A). In an effort to find a predictor of glans torsion, we studied the degree of median raphe deviation from midline at foreskin tip or coronal sulcus separately (Tables 3 and 4). If we consider as a starting point any degree of median raphe deviation greater than zero at the foreskin tip (Table 3A), in 27% of cases the glans had torsion. When the median
Table 2 Distribution of angle of median raphe deviation without glanular torsion (n Z 270) Angle ( )
Deviation of median raphe at foreskin tip
Deviation of median raphe at coronal sulcus
0e10 11e20 21e30 31e40 41e50 51e60 61e70 71e80 81e90
240 10 11 2 4 3 0 0 0
241 10 10 5 2 2 0 0 0
498
Sarkis P. E., S. Muthurajan
Table 3 Distribution of penile glanular torsion based on deviation of median raphe at foreskin tip: comparison between cut-off of 0 vs >0 in (A) and <10 vs >11 in (B) (A) Cut-off of median raphe deviation at foreskin tip fixed to 0 No glans torsion (only 0 )
With glans torsion (>5 )
Total
Based on median raphe deviation at foreskin tip Cut-off 0 (%)
Cut-off 10 (%)
Cut-off 0 (%)
Cut-off 10 (%)
91 86 71
83 87 66
69 86 65
59 87 60
0 >0
233 (96%) 37 (29%)
9 (4%) 91 (71%)
242 128
Total
270 (73%)
100 (27%)
370
(B) Cut-off of median raphe deviation at foreskin tip fixed to 10 No glans torsion (<10 )
With glans torsion (>11 )
Table 5 Sensitivity, specificity and positive predictive value in predicting glans torsion based on two different cut-offs of median raphe deviation at foreskin tip or coronal sulcus
Total
Sensitivity Specificity Positive predictive value
Based on median raphe deviation at coronal sulcus
All differences are not significant; p > 0.05.
<10 >11
244 (94%) 38 (34%)
15 (6%) 73 (66%)
259 111
Total
282 (76%)
88 (24%)
370
raphe was deviated, 72% had torsion. The odds ratio (OR) of the association is 1.097 (1.074e1.118; p < 0.05). If we consider median raphe angle of deviation at coronal sulcus as an indicator at any degree greater than zero (Table 4A), in 27% of all cases the glans had torsion. When the median raphe was deviated, 65% had torsion. For further analysis, we considered the first 10 to have no clinical significance being possibly due to measurement imprecision. Then we fixed the cut-off of torsion to 10 and more, and analyzed data after recoding cases with <10 as without torsion. 1. If the angle of median raphe rotation at foreskin tip is considered as indicator (Table 3B), 24% of cases had
glans torsion, and 66% of the deviated raphe at foreskin tip cases had glans torsion, with OR of 1.076 (1.052e 1.1; p < 0.0001). 2. If the angle of median raphe rotation at coronal sulcus is considered as indicator (Table 4B), 24% of cases had glans torsion, and 60% of the cases with deviated raphe at this level had glans torsion, with OR of 1.028 (1.004e 1.053; p < 0.05). In an attempt to search for a predictive factor of glans torsion by using Spearman’s correlational coefficient, median raphe deviation at foreskin tip was found to have a better correlation with glans torsion than median raphe deviation at sulcus (0.727 vs 0.570; both significant at p < 0.01). Although the use of median raphe deviation at the foreskin tip as an indicator gives better sensitivity, the specificity of the two indicators is the same and no statistical significance was found between the positive predictive values (Table 5).
Discussion Table 4 Distribution of penile glanular torsion based on deviation of median raphe at coronal sulcus: comparison between cut-off of 0 vs >0 in (A) and <10 vs >11 in (B) (A) Cut-off of median raphe deviation at coronal sulcus fixed to 0 No glans torsion (only 0 )
With glans torsion (>5 )
Total
0 >0
233 (88%) 37 (35%)
31 (12%) 69 (65%)
264 106
Total
270 (73%)
100 (27%)
370
(B) Cut-off of median raphe deviation at coronal sulcus fixed to 10 No glans torsion (<10 )
With glans torsion (>11 )
Total
<10 >11
247 (87%) 35 (40%)
35 (13%) 53 (60%)
282 88
Total
282 (76%)
88 (24%)
370
Neonatal congenital penile glanular torsion is considered a rare disease with unknown etiology [4,9,10]. Its incidence and clinical importance may have been underestimated, with a paucity of data in the literature. The authors became interested in the subject because they encountered isolated penile torsion more frequently than expected, primarily in neonates. The study was done to assess the demographic characteristics of real isolated neonatal penile torsion of the glans and not cases secondary to other diseases. The long-term significance and relation with adult function are beyond the scope of this study, and are limited by the lack of long-term data on a large population. The most accepted definition is rotation of the penis in one plane, in contrast to the three-dimensional aspect used by Hseih et al. [2]. Isolated congenital penile torsion may exist in the same family, father to son [9]. The best indicator of penile torsion is the angle of glanular torsion. Glans accessibility is limited at birth due to being covered by the tight, stenotic foreskin, and so the condition is unrecognized until circumcision is performed or the foreskin is
Neonatal penile glanular torsion retracted. In our institution, circumcision is done systematically in the neonatal period for religious reasons. This gave us the opportunity to assess the incidence of the disease. To our knowledge, neonatal penile torsion has not previously been studied in all its aspects. In this survey, the incidence of isolated neonatal penile glanular torsion was 27.02% (95% CI: 22.2%e31.84%). This incidence is higher than found in the series of Ben Ari et al. [8]. The increase is due to the study design (wide range of inclusions), and to the systematic measurement of glans torsion after exposition in all circumcised newborns. Incidence may vary from 27% to 23.5% to 9.5% according to the cut-off angle of torsion at which the torsion is considered as true. The high incidence of this pathology highlights the importance of its neonatal diagnosis. The shaft is almost always rotated to the left, in a counterclockwise direction, as confirmed in this sample. The penile shaft is normal at the base of the penis [1,8]. Although the incidence of glanular torsion of >60 is rare, severe torsion has been found [8,11]. The position of the external median raphe may or not reflect the real angle of glanular torsion, as 4%e12% of cases with no deviation had effectively hidden glans torsion [1,8,10]. To our knowledge, a predictive factor has not been studied before. With a simple external index (median raphe), the surgeon can predict the hidden angle of glans torsion before doing the circumcision and splitting the prepuce. Our data demonstrate clearly that median raphe deviation at coronal sulcus and foreskin tip correlates to glans torsion. More specifically, torsion angle at the level of foreskin tip seems more strongly correlated to glans torsion than at the level of coronal sulcus, but without a significant difference. The change in coding of the torsion (cut-off) to a glans angle of 10 may achieve greater clinical significance and more useful association coefficients.
Limitations of the study 1. Hospital-based sample: with the advantage of being inexpensive and easy to recruit, there is the risk of selection bias related to the center that may create an effect on the patient category and pathology incidence. We do not think that the type of sample would have affected the data related to this pathology. 2. Clinical importance: questions regarding necessity for correction and the relation to adult isolated penile torsion are difficult to answer because of the need for similar data in adulthood. 3. Statistically, the correlation coefficient (association) seen between the rotation at tip of foreskin and glans is subject to the underlying condition of a linear association between the two variables.
499
Conclusion The incidence of isolated neonatal penile glanular torsion was 27% (95% CI: 22.2e31.84%). The majority (87%) were between 10 and 20 . Torsion was to the left in 99% of cases. Isolated neonatal penile torsion is more common than reported. The median raphe of the penis may be normal and mask unexpected glans torsion. Median raphe torsion at the level of foreskin tip can be used as an indicator for glans torsion, with better association than torsion at the coronal sulcus. Clinical significance and relation to adult penile torsion need long-term follow-up data.
Acknowledgments The authors thank Dr Volker Rudat for his critical discussion and the nursing team of the hospital for their support during the realization of the article. Special thanks to Mrs N. Robles, Miss N. Emit, Miss I. Al-Nasser for manual information processing of data.
References [1] Elder JS. Male genital anomalies. Lateral or dorsal curvature of the penis. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, editors. Campbell’s urology. 8th ed. Philadephia: W.B. Saunders; 2002. p. 2342e3. [2] Hseih J-T, Wong W-Y, Chen J, Chang H-J, Liu S-P. Congenital isolated penile torsion in adults. Untwist with placation. Urology 2002;59:438e40. [3] Peter FC, John PM. Penile torsion repair using dorsal dartos flap rotation. J Urol 2004;171:1902e4. [4] Jordan GH, Schlossberg SM. Surgery of the penis, curvatures of the penis. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, editors. Campbell’s urology. 8th ed. Philadephia: W.B. Saunders; 2002. p. 3942e3. [5] Redman JF. Rare penile anomalies presenting with complication of circumcision. Urology 1988 Aug;32:130e2 [Abstract]. [6] Zhou L, Mei H, Hwang AH, Xie H-W, Hardy BE. Penile torsion repair by suturing tunica albuginea to the pubic periosteum. J Pediatr Surg 2006;41:E7e9. [7] Asif R, Allan MS, Paul H. Penile necrosis secondary to spontaneous penile torsion. J Urol 2001;166:215. [8] Ben Ari J, Merlob P, Mimouni F, Reisner SH. Characteristics of the male genitalia in the newborn. J Urol 1985;134:521e2. [9] Paxson Jr CL, Corriere Jr JN, Morriss Jr FH, Adcock III EW. Congenital torsion of the penis in fathereson pairs. J Urol 1977; 118:881. [10] Pomerantz P, Hanna M, Levitt S, Kogan S. Isolated torsion of penis. Report of 6 cases. Urology 1978;11:37. [11] Redman JF. Torsion of the penis: a gross anatomic study. J Urol 1983;130:316.
Incidence and predictive factors of isolated neonatal penile glanular torsion Sarkis Pierrot E.*,1, Sadasivam Muthurajan1,2 Pediatric Surgery Section, Saad Specialist Hospital, Hammoud Street, P.O. Box: 30353, Al-Khobar 31952, Saudi Arabia Received 23 January 2007; accepted 7 March 2007 Available online 5 July 2007
KEYWORDS Penile malformations; Neonatal penile torsion; Median raphe deviation; Penile glanular torsion
Abstract Purpose: To determine the incidence of isolated neonatal penile glanular torsion, describe the basic characteristics, and explore the relationship between foreskin and glans torsion. Methodology: A prospective survey was conducted of all male newborns admitted to nursery after delivery, or neonates less than 3 months presenting for circumcision. Cases with associated genital malformations were excluded. Results: The incidence of isolated neonatal penile torsion was 27% (95% CI: 22.2%e31.84%), to the left in 99% of cases. In 3.5% of cases, the penis had an angle <10 , and 9.5% >20 . Using Spearman’s correlational coefficient, deviation of penile raphe from the midline at the foreskin tip had a better correlation with glans torsion than deviation of raphe at the coronal sulcus (0.727 vs 0.570; both significant at p < 0.01). Conclusion: Isolated neonatal penile torsion is more common than reported. The median raphe of the penis may be normal and mask unexpected glans torsion. Median raphe torsion at foreskin tip can be used as a predictor for glans torsion. Clinical significance and relation to adult penile torsion are beyond the scope of the study. ª 2007 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.
Introduction Penile torsion is a two- or three-dimensional rotation of the penis at the level of the shaft and/or the glans [1,2]. Pediatric surgeons frequently encounter penile torsion at the neonatal stage in countries where circumcision is routinely performed after delivery. The condition can be diagnosed * Corresponding author. Tel.: þ966 501617127/þ966 38014810. E-mail addresses: [email protected], psarkis@saad. com.sa (Sarkis P.E.), [email protected] (Sadasivam M.). 1 Tel.:þ966 3882666; fax: þ96638823334. 2 Tel.: þ966 38014807.
following delivery, or later during childhood. Neonatal penile torsion can be isolated or associated with male genital malformations [3], and the etiology is different from the adult type [4e7]. There are few publications on isolated neonatal penile torsion, and the clinical characteristics are not well known. The real incidence is difficult to evaluate for many reasons. First, isolated torsion should be separated from secondary cases due to associated penile anomalies like hypospadias [3]. Second, the deviation of the penile median raphe from the midline may or may not reflect the real angle of glanular torsion. Penile torsion should be assessed by measuring the angle of glanular torsion not foreskin alone.
1477-5131/$30 ª 2007 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company. doi:10.1016/j.jpurol.2007.03.002
496
Sarkis P. E., S. Muthurajan
We encountered glanular penile torsion more frequently than is reported [8]. The aims of this study were to determine the following. 1. The incidence of isolated neonatal penile glanular torsion, and its descriptive characteristics, in a country (Saudi Arabia) where circumcision is almost systematic at the neonatal stage mainly for ritual or religious reasons. 2. The correlation between glans torsion and foreskin torsion. 3. The possibility of using median raphe deviation or torsion as a predictor of glans torsion.
Patient and methods Between July and December 2005, a prospective survey was conducted at the Saad Specialist Hospital, AL-Khobar, KSA, of all newborn candidates for circumcision (Fig. 1).
Exclusion criteria 1. Newborn transferred to neonatal intensive care unit or with major medical pathology. 2. Refusal of neonatal circumcision by parents. 3. Insufficient data. 4. All secondary cases: defined as associated major congenital anomaly of genital organs, or coexistence of hypospadias, chordee or lateral curvature of penis.
Sampling and recruitment This was a hospital-based sample. A consecutive sampling technique was chosen because of absence of any seasonal factor or time-dependent trends, or any source of systematic error. Institutional review board and ethical committee approval was obtained for the study.
Strategies to enhance precision Inclusion criteria 1. Every male newborn admitted to nursery after delivery. 2. Or neonate less than 3 months of age, presenting to pediatric surgery outpatient clinic for circumcision. 3. Newborns with unilateral undescended testis, unilateral or bilateral hydrocele were not excluded.
Candidates for circumcision: 1- All male newborn delivered in SSH 2- OPD males <3 months old Between 15 July and 30 November 2005 (470 male/859 live births)
1st step exclusion criteria (83 boys): 1- Discharged against medical advice (24) or early discharge post delivery (18) 2- Transfer to NICU or any major medical contraindication (26) 3- Parents refuse circumcision (15)
All circumcisions and measurement were done by two surgeons personally. To ensure internal validity, and consistency between observers, the methodology of measurement has been standardized and recorded in writing. Sessions for practice of the technique were performed in the presence of both authors for the first 10 cases, and repeated every first week of each month.
Outcome of interest (endpoint) Isolated congenital neonatal penile torsion was measured by the degree of torsion at the level of the glans, which was compared to the angle of deviation of the median raphe at two different levels: coronal sulcus and foreskin tip (Figs. 2 and 3). For the purpose of this study, angles are grouped by intervals of 10 . To measure the angles with precision and
Eligible for the study (387 boys)
2nd step exclusion criteria (17 boys): 1- No or insufficient data (4) 2- Major congenital anomalies: ambiguous genitalia (1)… 3- Coronal or severe hypospadias (1) 4- Chordee (1) 5- Lateral curvature minor or severe (4) 6- Webbed penis (5) 7- No raphe apparent (1)
Final study population = 370 boys 1- Normal = 270 2- Isolated neonatal penile glanular torsion= 100
Figure 1 Flowchart of study population for isolated neonatal penile glanular torsion with inclusion and exclusion criteria. SSH, Saad Specialist Hospital; OPD, outpatient department; NICU, neonatal intensive care unit.
Level of measurement of median raphe deviation at foreskin tip
Level of measurement of median raphe deviation at coronal sulcus
Figure 2 Ventral median raphe highlighted to improve visualization and accuracy of measurement, and indication of the three levels of measurement (coronal sulcus, foreskin tip, and glans).
Neonatal penile glanular torsion
497 Table 1 Distribution of angle of median raphe deviation and glanular torsion (n Z 370)
Angle of torsion of meatus indicating the torsion of glans
Level of measurement of median raphe deviation at foreskin tip Level of measurement of median raphe deviation at coronal sulcus
Figure 3 Ventral median raphe highlighted to improve visualization and accuracy of measurement, and indication of the three levels of measurement (coronal sulcus, foreskin tip, and glans).
accuracy, the following technical adjustments were undertaken. 1. A small regular protractor was used, with minor modification for a better adjustment around the penile shaft. 2. Torsion angle of the foreskin: Before injection of penile block analgesia, the midline ventral penile raphe was highlighted to improve its visibility. The penis was put in a straight upright position by digital pressure on the dorsal root of the penis avoiding torsion. The angle of median raphe deviation from the midline at the level of the coronal sulcus and the tip was measured. 3. Torsion angle of the glans: The foreskin was split dorsally at the midline, and adhesiolysis of the glans was completed, as a first step for circumcision. Before any manipulation, the degree of torsion of the glans was measured at the level of the urethral meatus with a sterile protractor. For a descriptive study with 95% CI of 10%, and based on the possible incidence of neonatal penile torsion calculated from a pilot unpublished internal study, a sample of 330 male newborns was needed; this was achieved in 6 months. Statistical analysis was done using SPSS 14. Spearman nonparametric correlation coefficient between continuous or ordinal variables not normally distributed was calculated. To quantify the level of association and prediction of torsion, logistic regression analysis between the dependent ‘glans torsion’ and the independent angle of median raphe torsion at foreskin tip and coronal sulcus was done, and positive predictive values (compared by Chi-square test) were calculated. Significance level was fixed to two-tailed test with p < 0.05.
Results Data were collected on 387 male infant candidates for circumcision, of whom 105 had neonatal penile glanular torsion with an incidence of 27.4% (95% CI: 23%e31.8%). After exclusion of 17 files because of coexistence of genital
Angle ( )
Deviation of median raphe at coronal sulcus
Deviation of median raphe at foreskin tip
Torsion of glans
0 1e10 11e20 21e30 31e40 41e50 51e60 61e70 71e80 81e90
264 18 24 28 8 14 10 1 2 1
242 17 35 30 6 19 12 4 0 5
270 13 52 23 3 8 1 0 0 0
pathology (Fig. 1), the incidence of isolated neonatal penile torsion based on glans torsion was 27.02% (95% CI: 22.2%e 31.84%). Torsion was to the left in 99% of cases, in a counterclockwise direction, and in two cases to the right (2/370; <1%). Another case of right penile torsion was associated with right curvature and excluded from the study. Of the 370 included cases, 25 infants were non-Saudis. In the group of Saudi male newborns, penile glanular torsion was present in 82 cases (24%), with 10 out of 82 at less than 10 . The distribution of torsion at the three measured levels is shown for the whole population (Table 1) and for cases without glanular torsion (Table 2). No glans torsion exceeded 60 . There was special interest in the relationship between deviation of the median raphe and glans torsion. Even when there was no deviation from the midline, some glans showed unexpected torsion: in 4%e12%, depending on whether the principal indicator was fixed as the angle of deviation of median raphe at the coronal sulcus or foreskin tip (Tables 3A and 4A). In an effort to find a predictor of glans torsion, we studied the degree of median raphe deviation from midline at foreskin tip or coronal sulcus separately (Tables 3 and 4). If we consider as a starting point any degree of median raphe deviation greater than zero at the foreskin tip (Table 3A), in 27% of cases the glans had torsion. When the median
Table 2 Distribution of angle of median raphe deviation without glanular torsion (n Z 270) Angle ( )
Deviation of median raphe at foreskin tip
Deviation of median raphe at coronal sulcus
0e10 11e20 21e30 31e40 41e50 51e60 61e70 71e80 81e90
240 10 11 2 4 3 0 0 0
241 10 10 5 2 2 0 0 0
498
Sarkis P. E., S. Muthurajan
Table 3 Distribution of penile glanular torsion based on deviation of median raphe at foreskin tip: comparison between cut-off of 0 vs >0 in (A) and <10 vs >11 in (B) (A) Cut-off of median raphe deviation at foreskin tip fixed to 0 No glans torsion (only 0 )
With glans torsion (>5 )
Total
Based on median raphe deviation at foreskin tip Cut-off 0 (%)
Cut-off 10 (%)
Cut-off 0 (%)
Cut-off 10 (%)
91 86 71
83 87 66
69 86 65
59 87 60
0 >0
233 (96%) 37 (29%)
9 (4%) 91 (71%)
242 128
Total
270 (73%)
100 (27%)
370
(B) Cut-off of median raphe deviation at foreskin tip fixed to 10 No glans torsion (<10 )
With glans torsion (>11 )
Table 5 Sensitivity, specificity and positive predictive value in predicting glans torsion based on two different cut-offs of median raphe deviation at foreskin tip or coronal sulcus
Total
Sensitivity Specificity Positive predictive value
Based on median raphe deviation at coronal sulcus
All differences are not significant; p > 0.05.
<10 >11
244 (94%) 38 (34%)
15 (6%) 73 (66%)
259 111
Total
282 (76%)
88 (24%)
370
raphe was deviated, 72% had torsion. The odds ratio (OR) of the association is 1.097 (1.074e1.118; p < 0.05). If we consider median raphe angle of deviation at coronal sulcus as an indicator at any degree greater than zero (Table 4A), in 27% of all cases the glans had torsion. When the median raphe was deviated, 65% had torsion. For further analysis, we considered the first 10 to have no clinical significance being possibly due to measurement imprecision. Then we fixed the cut-off of torsion to 10 and more, and analyzed data after recoding cases with <10 as without torsion. 1. If the angle of median raphe rotation at foreskin tip is considered as indicator (Table 3B), 24% of cases had
glans torsion, and 66% of the deviated raphe at foreskin tip cases had glans torsion, with OR of 1.076 (1.052e 1.1; p < 0.0001). 2. If the angle of median raphe rotation at coronal sulcus is considered as indicator (Table 4B), 24% of cases had glans torsion, and 60% of the cases with deviated raphe at this level had glans torsion, with OR of 1.028 (1.004e 1.053; p < 0.05). In an attempt to search for a predictive factor of glans torsion by using Spearman’s correlational coefficient, median raphe deviation at foreskin tip was found to have a better correlation with glans torsion than median raphe deviation at sulcus (0.727 vs 0.570; both significant at p < 0.01). Although the use of median raphe deviation at the foreskin tip as an indicator gives better sensitivity, the specificity of the two indicators is the same and no statistical significance was found between the positive predictive values (Table 5).
Discussion Table 4 Distribution of penile glanular torsion based on deviation of median raphe at coronal sulcus: comparison between cut-off of 0 vs >0 in (A) and <10 vs >11 in (B) (A) Cut-off of median raphe deviation at coronal sulcus fixed to 0 No glans torsion (only 0 )
With glans torsion (>5 )
Total
0 >0
233 (88%) 37 (35%)
31 (12%) 69 (65%)
264 106
Total
270 (73%)
100 (27%)
370
(B) Cut-off of median raphe deviation at coronal sulcus fixed to 10 No glans torsion (<10 )
With glans torsion (>11 )
Total
<10 >11
247 (87%) 35 (40%)
35 (13%) 53 (60%)
282 88
Total
282 (76%)
88 (24%)
370
Neonatal congenital penile glanular torsion is considered a rare disease with unknown etiology [4,9,10]. Its incidence and clinical importance may have been underestimated, with a paucity of data in the literature. The authors became interested in the subject because they encountered isolated penile torsion more frequently than expected, primarily in neonates. The study was done to assess the demographic characteristics of real isolated neonatal penile torsion of the glans and not cases secondary to other diseases. The long-term significance and relation with adult function are beyond the scope of this study, and are limited by the lack of long-term data on a large population. The most accepted definition is rotation of the penis in one plane, in contrast to the three-dimensional aspect used by Hseih et al. [2]. Isolated congenital penile torsion may exist in the same family, father to son [9]. The best indicator of penile torsion is the angle of glanular torsion. Glans accessibility is limited at birth due to being covered by the tight, stenotic foreskin, and so the condition is unrecognized until circumcision is performed or the foreskin is
Neonatal penile glanular torsion retracted. In our institution, circumcision is done systematically in the neonatal period for religious reasons. This gave us the opportunity to assess the incidence of the disease. To our knowledge, neonatal penile torsion has not previously been studied in all its aspects. In this survey, the incidence of isolated neonatal penile glanular torsion was 27.02% (95% CI: 22.2%e31.84%). This incidence is higher than found in the series of Ben Ari et al. [8]. The increase is due to the study design (wide range of inclusions), and to the systematic measurement of glans torsion after exposition in all circumcised newborns. Incidence may vary from 27% to 23.5% to 9.5% according to the cut-off angle of torsion at which the torsion is considered as true. The high incidence of this pathology highlights the importance of its neonatal diagnosis. The shaft is almost always rotated to the left, in a counterclockwise direction, as confirmed in this sample. The penile shaft is normal at the base of the penis [1,8]. Although the incidence of glanular torsion of >60 is rare, severe torsion has been found [8,11]. The position of the external median raphe may or not reflect the real angle of glanular torsion, as 4%e12% of cases with no deviation had effectively hidden glans torsion [1,8,10]. To our knowledge, a predictive factor has not been studied before. With a simple external index (median raphe), the surgeon can predict the hidden angle of glans torsion before doing the circumcision and splitting the prepuce. Our data demonstrate clearly that median raphe deviation at coronal sulcus and foreskin tip correlates to glans torsion. More specifically, torsion angle at the level of foreskin tip seems more strongly correlated to glans torsion than at the level of coronal sulcus, but without a significant difference. The change in coding of the torsion (cut-off) to a glans angle of 10 may achieve greater clinical significance and more useful association coefficients.
Limitations of the study 1. Hospital-based sample: with the advantage of being inexpensive and easy to recruit, there is the risk of selection bias related to the center that may create an effect on the patient category and pathology incidence. We do not think that the type of sample would have affected the data related to this pathology. 2. Clinical importance: questions regarding necessity for correction and the relation to adult isolated penile torsion are difficult to answer because of the need for similar data in adulthood. 3. Statistically, the correlation coefficient (association) seen between the rotation at tip of foreskin and glans is subject to the underlying condition of a linear association between the two variables.
499
Conclusion The incidence of isolated neonatal penile glanular torsion was 27% (95% CI: 22.2e31.84%). The majority (87%) were between 10 and 20 . Torsion was to the left in 99% of cases. Isolated neonatal penile torsion is more common than reported. The median raphe of the penis may be normal and mask unexpected glans torsion. Median raphe torsion at the level of foreskin tip can be used as an indicator for glans torsion, with better association than torsion at the coronal sulcus. Clinical significance and relation to adult penile torsion need long-term follow-up data.
Acknowledgments The authors thank Dr Volker Rudat for his critical discussion and the nursing team of the hospital for their support during the realization of the article. Special thanks to Mrs N. Robles, Miss N. Emit, Miss I. Al-Nasser for manual information processing of data.
References [1] Elder JS. Male genital anomalies. Lateral or dorsal curvature of the penis. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, editors. Campbell’s urology. 8th ed. Philadephia: W.B. Saunders; 2002. p. 2342e3. [2] Hseih J-T, Wong W-Y, Chen J, Chang H-J, Liu S-P. Congenital isolated penile torsion in adults. Untwist with placation. Urology 2002;59:438e40. [3] Peter FC, John PM. Penile torsion repair using dorsal dartos flap rotation. J Urol 2004;171:1902e4. [4] Jordan GH, Schlossberg SM. Surgery of the penis, curvatures of the penis. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, editors. Campbell’s urology. 8th ed. Philadephia: W.B. Saunders; 2002. p. 3942e3. [5] Redman JF. Rare penile anomalies presenting with complication of circumcision. Urology 1988 Aug;32:130e2 [Abstract]. [6] Zhou L, Mei H, Hwang AH, Xie H-W, Hardy BE. Penile torsion repair by suturing tunica albuginea to the pubic periosteum. J Pediatr Surg 2006;41:E7e9. [7] Asif R, Allan MS, Paul H. Penile necrosis secondary to spontaneous penile torsion. J Urol 2001;166:215. [8] Ben Ari J, Merlob P, Mimouni F, Reisner SH. Characteristics of the male genitalia in the newborn. J Urol 1985;134:521e2. [9] Paxson Jr CL, Corriere Jr JN, Morriss Jr FH, Adcock III EW. Congenital torsion of the penis in fathereson pairs. J Urol 1977; 118:881. [10] Pomerantz P, Hanna M, Levitt S, Kogan S. Isolated torsion of penis. Report of 6 cases. Urology 1978;11:37. [11] Redman JF. Torsion of the penis: a gross anatomic study. J Urol 1983;130:316.