Introduction

Male circumcision (MC) is one of the oldest and most common surgical procedures in the world. It consists of the shaft skin and inner foreskin surgical removal to uncover the glans. The global prevalence of MC is estimated to be 38–39% [1]. Given the high rates of circumcision among Muslim and Jewish males, the prevalence of this procedure in the Middle East and North Africa reaches over 95%. In the USA, the prevalence of MC is 91% in White males, 76% in Black males, and 44% in Hispanic males [2]. In other English-speaking countries, the prevalence of MC is about 20–30%. In Asian and European countries, the prevalence ranges broadly and this is mainly dependent on the predominant religious beliefs [1]. MC can be performed at any age; it is most commonly performed in neonates, followed by infants and children with important differences in complication rates and related costs [3, 4].

Neonatal MC is a simple procedure, healing within 1 week, and when performed in a hospital setting by trained physicians, it has a low rate of adverse events [5]. The reasons for MC vary but most newborns are circumcised for religious or cultural reasons. There is considerable evidence of the health benefits connected to newborn male circumcision in preventing infectious and non-infectious diseases (i.e., prevention of urinary tract infections, acquisition of HIV, balanitis, paraphimosis, candidiasis, transmission of some sexually transmitted infections, and genital cancer) [5,6,7]. By contrast, some are less likely to endorse this procedure, and a debate is still ongoing in the medical community whether to recommend neonatal circumcision.

Three surgical devices are commonly used to perform MC: the Gomco clamp, the Plastibell device, and the Mogen clamp (Table 1).

Table 1 Surgical procedures for male circumcision
Full size table

MC performed without anesthesia is likely to be associated with intraoperative and post-operative pain. Therefore, MC should always be performed using anesthesia and post-operative analgesia [5, 8]. The relief of pain during medical procedures is a crucial goal for healthcare providers, and strong evidence shows that newborns can experience pain and that neuronal pathways are affected by painful stimuli as well as the future pain threshold [9, 10]. For the risk of neurological damage secondary to the use of general anesthetics, general anesthesia for MC in newborns should be avoided [11]. The use of local anesthesia for MC has proven to be a safe and effective way to prevent procedural pain both in neonates and in older children [5, 12,13,14,15].

For many years, physicians have tried to search for analgesic methods to perform MC safely but a gold standard for this practice has not yet been found. The aim of the present study is to systematically review the studies on MC pain management published during the past two decades.

Methods

We included in our search the original studies, published in the last two decades, comparing the effects of different types of anesthesia/analgesia for neonatal MC or the effects of different surgical techniques for MC on pain levels. We included only studies in which pain level was assessed using specific pain scales or physiologic and behavioral responses to pain. We considered only the studies in which MC was performed during the neonatal period. Only studies written in English were included in the analysis. Reviews, meta-analyses, commentaries, and quality improvement projects were excluded. We retrieved papers from PubMed, MEDLINE, and Cochrane databases using the following combining keywords and MeSH terms: pain, anesthesia, analgesia, newborn, and male circumcision. We consensually decided to use the aforementioned keywords altogether, aiming to increase the possibility of retrieving well-matched studies. Additional sources used to identify studies included the reference lists of relevant articles and Google Scholar. The final literature search was performed in January 2020 initially with the scanning of titles and abstracts for inclusion and then with the assessment of full-text articles performed by two authors independently (CVB and SR) in order to reduce the possibility of rejecting relevant reports. The study selection process is displayed using the PRISMA flow diagram [16] (Fig. 1). Data were extracted by one author (SR) using a data extraction form, which was independently checked by a second researcher (CVB). Data extracted included the year of publication, the study type, the sample size, the analgesic methods applied, the pain scale used, the surgical technique utilized, and the main results of each study. The data collected were summarized descriptively.

Fig. 1
figure 1

PRISMA 2009 flow diagram

Full size image

Results

Forty-seven papers published from the year 2000 were retrieved after initial database search using the aforementioned combined keywords. Forty-five were screened after duplicate removal. Sixteen of them were excluded after the initial scanning of titles and abstract, and eight were excluded after full-text analysis because they did not fulfill our inclusion criteria. Twenty-one studies met the inclusion criteria and were included in the analysis.

Seventeen articles compared the effect of different analgesic methods in newborns’ MC [17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33] while 4 articles compared the effects of different surgical techniques on pain levels [35,36,37,38]. Thirteen studies enrolled more than 20 patients for each study group [17, 18, 20, 21, 23, 24, 27,28,29,30,31,32,33].

Table 2 displays, in chronological order, a summary of the key features and findings of each of the articles that met the inclusion criteria. Table 3 shows the articles comparing different surgical techniques.

Table 2 Comparison of analgesic methods for male circumcision
Full size table
Table 3 Comparison of surgical techniques for male circumcision
Full size table

The main pharmacological methods used in the retrieved trials were as follows: EMLA cream which is an eutectic mixture of local anesthetics, with 2.5% lidocaine and 2.5% prilocaine producing dermal analgesia when applied as a topical cream on the skin 60–90 minutes before the procedure [5, 22, 26, 31, 32]; dorsal penile nerve block (DPNB) which involves regional anesthesia often obtained with 0.4 ml of 1% lidocaine or with bupivacaine injected at the base of the penis at the 10:00 and 2:00 positions [5, 17, 21, 23,24,25]; subcutaneous penile ring block (RB) which consists of the injection of 0.8 ml of 1% lidocaine in a circumferential ring around either the midshaft or at the level of the corona [5, 30, 32].

Other pharmacological interventions used were caudal block with bupivacaine [29], topical lidocaine cream (LMX4) containing 4% lidocaine [23], and acetaminophen/tylenol administered at the dosage of 10 mg/kg [18, 20, 24]. Among the non-pharmacological measures for pain control, the most commonly used were breast milk ad libitum during the procedure [28], sucrose solution (24–25% sucrose) [22, 30,31,32], dextrose solution (50% dextrose) [17], non-nutritive sucking (NNS) provided in the form of a gloved finger [24], and audio-stimulation with music (lullabies or repetitive rhythms) [19].

The ways used to assess pain response in neonates were both specific pain scales and physiological and behavioral responses to pain. NIPS (Neonatal Infant Pain Scale) was the most commonly used. Other methods to establish pain level were PIPP (premature infant pain profile), FLACC (Face, Legs, Activity, Cry, Consolability scale), NFCS (Neonatal Facial Coding System), N-PASS (Neonatal Pain, Agitation, and Sedation Scale), MBPS (Modified Behavioral Pain Scale), crying time, and the assessment of vital signs such as heart rate, oxygen saturation, and respiratory rate changes.

Ten articles, with more than 20 patients enrolled for each group, recorded pain levels during MC using validated pain scales [18, 20, 21, 24, 27,28,29,30,31,32].

Dorsal penile nerve block

Ten papers compared DPNB with other anesthetic techniques [17, 21,22,23,24,25,26,27, 32, 33].

In two papers, DPNB was performed using bupivacaine [21, 25]; in one, DPNB with bupivacaine was proven more effective than DPNB with lidocaine [25]. DPNB with lidocaine was used for all the other studies [17, 22,23,24, 26, 27, 32, 33]. In 6 studies, DPNB was found to be more effective than the alternative measures used [17, 21, 22, 26, 32, 33]. In 4 studies, DPNB was combined with other forms of analgesia such as sucrose, EMLA, tylenol, and non-nutritive sucking. In all, the combined use of DPNB and additional pain-relieving methods proved to be effective [22, 24, 25, 32].

EMLA

Nine studies focused on the effects of EMLA topical cream in producing analgesia during MC comparing it with other methods [19, 21,22,23, 26, 27, 31,32,33]. One study showed higher pain scores when EMLA cream was used as a sole analgesic method [32], and in the papers by Garry et al. and by Modekewe et al., EMLA was proven inferior to DPNB [26, 33]. Two studies demonstrated that the effectiveness of EMLA was higher when combined with other analgesic strategies, either pharmacological or non-pharmacological [31, 32]. Three studies found no significant differences in the comparison of EMLA with other analgesic strategies, namely DPNB and lidocaine cream [21, 23, 27]. In one study, EMLA proved to be more effective when combined with music [19]. An older study by Russell et al., in which MC was performer with Plastibell technique, reported a complete alleviation of pain with the application of EMLA 1 h prior to the procedure [34]. Another study by Taddio et al. in 1997 recorded behavioral (facial activity and time spent crying) and physiologic (heart rate and blood pressure) responses during MC performed with Gomco clamp, and demonstrated that EMLA is efficacious and safe for the prevention of pain from circumcision in neonates [35].

Ring block

Four papers analyzed the effectiveness of ring block anesthesia either alone or combined with other strategies [22, 28, 30, 32]. RB combined with oral sucrose was found to be more effective than RB alone [22, 30]. In one study, RB combined with EMLA and sucrose was significantly more effective than both EMLA + sucrose and DPNB + EMLA + sucrose [32]. In one study, RB along with breast milk and/or 20% sucrose resulted in painless circumcision for neonates under 2 weeks old, with a NIPS score of 0 to 2 [28].

Caudal block

One study assessed the effectiveness of caudal block performed either with low-volume high-concentration bupivacaine (0.375%) or standard dose of bupivacaine (0.25%). Block onset time and NIPS score were identical between the two groups demonstrating the effectiveness of high-concentration bupivacaine [29].

Acetaminophen

Two studies compared the use of acetaminophen vs. placebo to obtain analgesia during MC [18, 20]. One study found no difference between the two groups [18]; by contrast, the paper by Malnory et al. showed a stronger analgesia in newborns receiving acetaminophen [20].

Lidocaine cream

Two papers evaluated the use of lidocaine 4% cream comparing it with lidocaine 2.5% cream, EMLA, and DPNB. Pain response was evaluated using the changes in vital signs and no significant modifications in heart rate were found with different analgesic methods. Mean respiratory rate during the procedure was lower with lidocaine 4% cream and with DPNB compared with EMLA [23, 27].

Sucrose

Results from multiple studies discourage the use of sucrose alone due to the lack of its analgesic effect [17, 22, 30, 31].

MC surgical techniques

Thirteen out of 17 studies disclosed which surgical techniques were used for MC, Gomco being the most commonly utilized. Four papers compared the differences in pain profile with the use of different surgical techniques [36,37,38,39]. All studies demonstrated that MC performed with Mogen clamp was faster. Three papers compared Mogen clamp with Gomco clamp, demonstrating that Mogen clamp causes less pain especially when combined with preoperative analgesia (either with DPNB plus EMLA and acetaminophen, RB plus sucrose, or EMLA plus sucrose) [36, 37, 39]. Another study compared Mogen clamp with Plastibell procedure, using DPNB + dextrose as analgesia, finding no significant differences in terms of crying time [38].

Discussion

It was once thought that newborns were unable to feel pain and that painful sensations experienced during the neonatal period would have easily been forgotten by them. It is now known, however, that newborns do feel pain and there is evidence that this can have physiological and behavioral consequences. Newborns are strongly susceptible to painful stimuli because, while ascending neural pathways for nociception are well-developed by the second trimester of pregnancy, the modulatory descending pathways are immature, even in term newborns [40,41,42]. There is ongoing research regarding the effects of prolonged or repetitive painful stimuli on newborns’ neurodevelopment, provoking a decreased brain growth in the frontal and parietal lobes and alterations in organization of neuronal connections in the temporal lobes [9, 43, 44]. Taddio et al. demonstrated a stronger pain response to routine vaccination in infants circumcised without anesthesia in comparison with uncircumcised infants or circumcised babies treated with topical anesthesia; they argued that MC without anesthetic may induce changes in infant pain behavior because of alterations in the neural processing of painful stimuli [10]. By contrast, a recent analysis by O’Callahan et al. demonstrated that newborns circumcised with proper analgesia (24%sucrose plus local anesthesia with lidocaine or DPNB and acetaminophen) showed no differences in the rate of exclusive breastfeeding during the initial hospitalization compared with non-circumcised newborns [45]. Therefore, it is imperative for clinicians to attempt all the possible strategies to reduce neonatal pain, and since MC is an invasive painful procedure, performing it without anesthesia appears to be contraindicated [8].

The studies we retrieved displayed heterogeneity in terms of sample size, pain scales, analgesic methods, and surgical techniques used. In some studies, the surgical details were not specified, and others used non-specific methods for pain evaluation (e.g., crying time, heart rate, respiratory rate). As a consequence and limitation of the study, it was not possible to perform a meta-analysis of different studies. Analyzing the studies with more than 20 patients for each study group and that used a validated pain scale, it was clear that the best results were obtained with pharmacologic local anesthesia and in particular the combination of pharmacological and non-pharmacological methods [20, 21, 24, 27,28,29,30,31,32]. As demonstrated by South et al., adding non-nutritive sucking to a combination of DPNB and tylenol resulted in lower pain scores compared with the use of pharmacological methods alone [24].

The value of combining non-pharmacological analgesic methods with DPNB has been demonstrated not only in newborns but also in older infants [12]. Similarly, by adding sucrose to other analgesic methods, either EMLA, DPRB, or RB, the result was a minimal pain level during MC [30,31,32]. These findings are in accordance with the work by Stevens at al. demonstrating the value of sucrose administration during painful procedures in neonates [46].

Only one study compared DPNB with RB, both associated with EMLA and sucrose, demonstrating a better analgesic effect when the combination of RB + EMLA + sucrose was used [32]. This finding confirmed the results of an older study by Lander et al. showing that RB is superior to DPNB [47].

Regarding surgical techniques, the general belief is that Mogen technique is shorter in terms of duration compared with Gomco or Plastibell [36,37,38,39]. A large study by Sinkey et al. suggests that Mogen technique is also less painful than Gomco [39]. However, in the majority of trials comparing analgesic methods that were considered in this review, MC was performed using Gomco. One reason for this might be the fact that the choice of the surgical technique is variable between centers, and there seems to be a trend toward the use of Gomco in Western countries’ in-hospital settings [48, 49].

Two studies focused mainly on post-operative analgesia showing lower pain levels and longer-lasting analgesia with the use of bupivacaine as an anesthetic agent during MC [25, 29].

Despite the great variability that undoubtedly exists regarding the provision of analgesia during MC, the use of a combined pharmacological and non-pharmacological approach could be considered safe and effective. As we stated in our previous work on this topic, we are convinced that a possible valuable adjunct to the current perspective might be the integration of sensorial saturation (SS) as an aid to pharmacological analgesia for MC [50]. SS is a validated method to reduce newborns’ stress response to nociception and is based on the so-called gate-control theory by which the brainstem can filter and reduce the transmission of pain to the brain if the person is concentrated on something else. In this way, by attracting babies’ attention with positive stimuli (tactile, auditory, gustatory, and visual), it is possible to reduce and even nullify the perception of painful stimuli [51].

In conclusion, since a standardized protocol for analgesia during MC is yet to be determined, we encourage more research in this field as a possible implementation of current clinical practice, in order to ameliorate the current strategies so that newborns’ MC might eventually become a painless procedure. Given the proven effects of SS, a possible strategic direction to pursue in the field of MC analgesia could be the combination of pharmacological methods with SS, possibly potentiating the already proven positive effects of sucrose in conjunction with anesthetic drugs.