L.J.S. and J.Q.Z. contributed equally to this work
To investigate the early and delayed effects of cavernous nerve electrocautery injury (CNEI) in a rat model, with the expectation that this model could be used to test rehabilitation therapies for erectile dysfunction (ED) after radical prostatectomy (RP).
Materials and Methods
In all, 30 male Sprague-Dawley rats were randomly divided equally into two groups (15 per group). The control group received CNs exposure surgery only and the experimental group received bilateral CNEI.
At 1, 4 and 16 weeks after surgery (five rats at each time point), the ratio of maximal intracavernosal pressure (ICP) to mean arterial pressure (MAP) was measured in the two groups. Neurofilament expression in the dorsal penile nerves was assessed by immunofluorescent staining and Masson's trichrome staining was used to assess the smooth muscle to collagen ratio in both groups.
At the 1-week follow-up, the mean ICP/MAP was significantly lower in the CNEI group compared with the control group, at 9.94% vs 70.06% (P < 0.05).
The mean ICP/MAP in the CNEI group was substantially increased at the 4- (35.97%) and 16-week (37.11%) follow-ups compared with the 1-week follow-up (P < 0.05).
At all three follow-up time points, the CNEI group had significantly decreased neurofilament staining compared with the control group (P < 0.05). Also, neurofilament expressions in the CNEI group at both 4 and 16 weeks were significantly higher than that at 1 week (P < 0.05), but there was no difference between 4 and 16 weeks (P > 0.05).
The smooth muscle to collagen ratio in the CNEI group was significantly lower than in the control group at the 4- and 16-week follow-ups (P < 0.05), and the ratio at 16 weeks was further reduced compared with that at 4 weeks (P < 0.05).
In the CNEI rat model, we found the damaging effects of CNEI were accompanied by a decline in ICP, reduced numbers of nerve fibres in the dorsal penile nerve, and exacerbated fibrosis in the corpus cavernosum.
This may provide a basis for studying potential preventative measures or treatment strategies to ameliorate ED caused by CNEI during RP.
Neuroprotective interventions have recently gained much attention as approaches to preserve erectile function during radical prostatectomy (RP). Despite technical and anatomical advances, erectile dysfunction (ED) remains a major complication of nerve-sparing RP (NSRP) . It is well known that the cause of ED after RP is mainly secondary to cavernous nerve (CN) injury. In many instances during RP the CNs may be inadvertently damaged by manipulation, separation, ligation, haemostasis, etc. Devices, e.g. monopolar and bipolar electrocautery, are routinely used to achieve haemostasis and facilitate dissection during RP. These devices produce potential thermal damage to the CNs, especially when the thermal energy is used in the posterolateral region of the prostate to release the neurovascular bundle. For example, a spot cautery on the vessels can result in a ‘heat sink’ injury to the adjacent CN fibres. Thus, electrocautery injury (EI) to the CNs is a potential cause of ED and as such cannot be ignored.
CN injury can be caused by various conditions, e.g. stretching, crushing, freezing, transecting, and excising the CNs, as well as unilateral vs bilateral CNs injury, which have been described in rat models [2, 3]. These models have been developed for use by researchers to mimic the damage and degeneration associated with RP. However, to our knowledge, no rat model to date has evaluated the effects of EI to the CNs. The present study was designed to investigate the early and delayed effects of CN EI (CNEI), with the expectation that this rat model could be used in the future to test rehabilitation therapies aimed at treating ED.
Materials and Methods
In all, 30 age-matched male Sprague-Dawley rats (300–350 g) were randomly equally divided into two groups (15 per group). Group 1 had surgery only (identification of the bilateral CNs), while Group 2 had surgery and bilateral CNEI. All experiments were carried out in accordance with Shanghai Sixth People's Hospital guidelines for animal care and use. All rats were housed with a 12-h light/dark cycle in a temperature controlled facility during the entire experimental period.
The surgical procedure was performed under anaesthesia with sodium pentobarbital (40 mg/kg, i.p.). The rats were fixed in the dorsal decubitus position. After establishing unresponsiveness to tail pinch reflex, hair in the low abdomen was shaved and a lower midline abdominal incision exposed the prostate gland and seminal vesicle. The major pelvic ganglion (MPG) was identified on both sides of the prostate under an operating microscope (×10), with its inflows from the hypogastric and pelvic nerves and CN outflow. Among the efferent nerves, the largest runs along the surface of the membranous urethra and is the main branch of the CNs (Fig. 1A,B). In the control group, the prostatic fascia and outside fat was dissected to expose the CNs, and no additional surgical manipulation was performed. For the CNEI group, after the exposure of the CNs as in the control group, the CNEI was induced by consistently applying a monopolar electrocautery tip connected to an electrocautery generator (Chunguang, HR-302, Wuhan, China) on the main branch of the CNs 5 mm distal to the MPG on both sides of the prostate. The monopolar electrocautery was performed for 1 s at 350 kHz and 15 W. Under the operating microscope, the CNs became grey and their continuity was maintained after EI (Fig. 1C). The laparotomy incisions were closed and the rats were placed under an infrared heat lamp to maintain body temperature until fully recovered.
Erectile Haemodynamic Evaluation
At either 1, 4 or 16 weeks after surgery, the rats (five rats/group at each time point) were anaesthetised again for the non-survival surgery. The methods have been described previously [4-6]. Through a repeat midline abdominal incision the CNs were exposed and isolated. The skin overlying the penis was incised and the crus of the penis identified. A stainless steel electrode connected to an electrical pulse stimulator (PowerLab, AD Instruments) was placed around the CNs at 1–2 mm proximal to the injury site, the intracavernosal pressure (ICP) was measured and recorded using a fine butterfly needle (23 G) with heparin (250 U/mL) that was attached to polyethylene (PE)-50 tubing, placed at the left crus of the penis. The electrical stimulations (1.5 mA with a pulse width of 5 ms, 20 Hz and 60 s) were conducted on either side separately. Systemic mean arterial pressure (MAP) was monitored via PE-50 tubing placed in the left carotid artery. The catheters were connected to a pressure transducer (AD Instruments, Castle Hill, New South Wales, Australia) for ICP and MAP monitoring. The ICP examination was performed by two ‘blinded’ observers. The ratio of maximal ICP to MAP was calculated to normalise for variations between rats.
After ICP examination, the rats were killed humanely. The middle part of the penile shafts were harvested and fixed for 24 h in 10% neutral buffered formalin, dehydrated, and embedded in paraffin for histology. Sections were cut (5 mm) and mounted using anti-fade mounting media (Vector Laboratories). For histological evaluation, immunofluorescence staining was performed for neurofilament (diluted 1:600; Abcam Inc.) in the dorsal nerves of the penis. Native tissue sections served as a positive control and tissue sections without primary antibodies served as the negative controls. Masson's trichrome staining was used to assess the smooth muscle to collagen ratio. Sections stained for neurofilament and with Masson's trichrome were evaluated by two independent and ‘blinded’ observers using a fluorescent microscope (Nikon Eclipse 80i, Japan). Computerised quantitative analysis was performed using Image-Pro Plus 5.1 software (Media Cybernetics, Inc., MD, USA) in 10 separate fields for each tissue sample.
The maximum ICP/MAP ratios, smooth muscle–collagen ratios, and neurofilament staining are reported as the mean (sd). Individual comparison between groups were analysed using the independent two-tailed t-tests and anova was used for comparing the three different follow-up time points; a P < 0.05 was considered to indicate statistical significance.
Erectile Haemodynamic Evaluation
There were no significant complications in either group after surgery. Erectile function was assessed by electrostimulation of the CNs at the designated time points after surgery. The ICP in the CNEI group elicited significantly decreased erectile responses compared with those in the control group (Fig. 2). At 1 week after surgery, the maximal ICP/MAP was significantly lower in the CNEI group than in the control group, at a mean (sd) of 9.94 (0.60)% vs 70.06 (3.64)% (P < 0.05). The maximal ICP/MAP in the CNEI group was substantially increased at the 4- (35.97 [4.49]%) and 16-week (37.11 [2.14]%) follow-ups compared with the 1-week follow-up (P < 0.05; Fig. 3). However, the maximal ICP/MAP at 16 weeks after surgery did not improve any further from that at 4 weeks in the CNEI group (P > 0.05), and was also obviously lower than in the control group [105.67 (4.76)%; P < 0.05; Fig. 3).
Neurofilament was analysed by immunofluorescence staining of the dorsal penile nerve (Fig. 4). Neurofilament expression was quantified as the relative percentage of neurofilament expression in the cross-section of the dorsal penile nerve fibers. At all three follow-up time points, the CNEI group had significantly decreased staining for neurofilament compared with the control group (P < 0.05). Also, neurofilament expression at both the 4- (47.10 [1.48]%) and 16-week (48.42 [0.60]%) follow-ups were significantly higher than that at the 1-week follow-up (37.28 [2.05]%; P < 0.05), but there was no difference between 4 and 16 weeks (P > 0.05; Fig. 5).
The corpus cavernosum was examined for the content of smooth muscle and collagen on slides stained with Masson's trichrome. The muscle content was decreased and collagen content was increased in the CNEI group as compared with the control group at both the 4- and 16-week follow-ups (Fig. 6). Using computerised quantitative analysis, the smooth muscle to collagen ratio in CNEI group (6.10 [0.25]%; 4.02 [0.34]%) was significantly lower than that in the control group (10.38 [0.85]%; 10.74 [0.48]%) at 4 and 16 weeks after surgery, respectively (P < 0.05), and the ratio at 16 weeks was further reduced compared with that at 4 weeks in the CNEI group (P < 0.05; Fig. 7).
To our knowledge, the present study is the first to prove the time-dependent damage of electrocautery induced CN injury in a rat model. With up to 16 weeks of follow-up, we found the damaging effects of CNEI were accompanied by a decline in ICP, reduced numbers of nerve fibres in the dorsal penile nerve, and exacerbated fibrosis in the corpus cavernosum when compared with the control group. The present CNEI model showed similar morphological and functional characteristics as models of CN crush injury [2, 3, 7].
Various methods of inducing injury in the CNs in rat models have been developed to mimic the damage associated with RP, including resection [8-11], transection [5, 12], crush [6, 13-15] and dissection [16, 17]. All these specific CN-injured animal models had been established to evaluate the pathological changes and response to different intraoperative manipulations. Resection and transaction methods result in discontinuation of the CNs, which represent severe CNs injury models and replicates that of non-NS RP. However, bilateral CN crush and dissection injury without interrupting the continuity of the CNs represent NSRP. With the hypothesis that EI on the CNs can lead to neuropathy and damage the structure of corpus cavernosum, we used CNEI to mimic the direct or indirect heat injury to the CNs during RP for energy sources such as electrocautery, ultrasonic shears and others, which are used at various steps for dissection or haemostasis.
The present study has shown that bilateral CN electrocautery in the rat abolishes erectile response to electrostimulation at 1 week after injury, and over time proceeding to 4 and 16 weeks after injury, the ICP level elicited had some degree of improvement, but was still lower than the control group. In addition, the nerve fibres in the dorsal nerve of the penis decreased to a minimum at 4 weeks after CNEI. Similarly, in a previous canine study, Ong et al.  suggested that the use of monopolar, bipolar, and harmonic energy for nerve dissection during RP appeared to cause significantly more injury to cavernosal function postoperatively. It is generally assumed, but has not been directly shown, that the mechanism of electrocautery injury on CNs may be related to high temperature, protein denaturation, ischaemic changes, or inflammatory responses.
The time-dependent recovery of penile haemodynamics and histological structure alterations in the present study were also reported in crush models . Kim et al.  reported that erectile function spontaneously restored to a significant degree at 6 months (ICP/MAP 63.3%) compared with 3 months (ICP/MAP 51.7%) after CN crush injury in a rat model. Mullerad et al.  reported that the ICP/MAP ratios at 3, 10, and 28 days after bilateral haemostat CN crush were 18%, 31%, and 32%, respectively. Using a mouse model, Jin et al.  reported that the ICP/MAP of the crush injury group gradually increased after 4 weeks and was significantly higher than that of the neurectomy group after 12 weeks. It is possible that CN regeneration may explain the gradual recovery of erectile function seen in these studies . We hypothesise that, unlike the resection and transection injuries, electrocautery, crush and dissection injuries keep the CNs sheath relatively intact, which provides a potential nerve conduit for regenerating axons and re-innervation of the tissue.
In the present study, although erectile function began to recover at 4 weeks after CNEI, the smooth muscle/collagen ratio decreased in a sustained fashion even over the longer term (16 weeks). Similarly, histological studies in men after RP showed alterations in penile connective tissue. In cavernosal biopsies performed 1 year after RP, a decrease in trabecular elastic fibres and smooth muscle cells was noted, as well as an increase in collagen content with fibrosis, compared with preoperative biopsies . Using CN crush, transection and resection models, it has been shown that CN injury is coupled to apoptosis of penile smooth muscle cells and persistent penile fibrosis, which contribute to the development of venous leaking and corporal veno-occlusive dysfunction [10, 11, 23, 24]. We think that these irreversible changes perhaps partly explain why the ICP did not recover further from 4 to 16 weeks after CNEI in the present study.
In the present control group, with limited exposure of the CNs, there was an initial decrease in the ICP/MAP ratio at the 1-week follow-up, with recovery of the ratio evident after 4 weeks. We think that this transient decline in the ICP/MAP ratio may be associated with mild traction injury or damage to the blood supply of the CNs. This phenomenon has been reported by Mullerad et al. , who reported that the exposure group, which consisted of CNs visualisation only without direct manipulation, resulted in significant erectile impairment at 10 days after initial surgery compared with the control group. A similar response was also reported in a dissection model recently described by Yamashita et al. . These results reflect the real clinical course in which erectile function after NSRP decreases temporarily and recovers thereafter.
The present study has some limitations. Firstly, we merely present preliminary data showing changes in function and histology of structures after CNEI in a rat model. The exact mechanism of EI to the CNs is not clear. Secondly, the effect of electrocautery to the CNs should be time and temperature dependent, what sort of parameters are most representative of RP deserves further study. Thirdly, the method of electrocautery used in the present study cannot fully simulate the exact clinical situation because there are so many energy sources, including electrocautery, ultrasonic shears and laser, used during RP currently. A direct comparison of the different energy sources on injury of the CNs will provide the basis for the optimal choice of energy sources during RP. Lastly, stretch injury to the CNs is common during the manipulations of RP. So, in future studies, various stretch injuries combined with thermal injury or thermal-spread injury should be used to establish a suitable model of CN injury, which is more representative of NSRP.
In conclusion, to our knowledge, the present study is the first to prove damage to the CNs caused by electrocautery in a rat model. With up to 16 weeks follow-up, we found the damaging effects of CNEI were accompanied by a decline in ICP, reduced numbers of nerve fibres in the dorsal penile nerve, and exacerbated fibrosis in the corpus cavernosum. The present rat model of CNEI may provide a basis for studying potential preventative measures or treatment strategies to ameliorate ED caused by CNEI during RP.
This work was supported by the National Natural Science Foundation of China (30901489) and Shanghai Jiaotong University Morningstar Scholars Programme.