Electrical Stimulation in Sexual Dysfunction

Type

Mechanisms

Examples

Direct

Change in sexual desire from disease

Typically reduced, e.g., from high prolactin and anemia of chronic renal failure [1]. May be increased, e.g., from some brain disorders [2]

Disruptbn of genital response from disease

ED from multiple sclerosis [3], hypertension [4], orgasmic disorder from multiple sclerosis [5]

Disruptbn of genital response from surgery

Radical prostatectomy and ED [6], radical hysterectomy and reduced genital congestion/reduced lubrication [7], orgasmic disorder after radical vulvectomy [8]

Disruptbn of genital response from radiation

ED from vascular (and also likely nerve) damage after radiotherapy for prostate cancer [9]; vaginal stenosis and friability from radiation for pelvic cancer [10]

Dyspareunia and disruptbn of sexual desire and response from chemotherapy

Sudden ovarian failure after chemotherapy for breast cancer [11]; testicular failure after intensive chemotherapy for hematopoietic transplantation [12]

Disruptbn of sexual desire and response from antiandrogen treatment

GnRH therapy for prostate cancer [13]

Disruptbn of genital response from aromatase inhibitors

Loss of sexual genital sensitivity, and exacerbation of vaginal atrophy from aromatase inhibition post breast cancer [14]

Disruptbn of sexual desire and response from pain

Pain from any chronic condition is a potent sexual distraction

Disruption of sexual desire and response from nonhormonal medications

Narcotics can depress desire through gonadotropin suppressbn [15]; selective serotonin reuptake inhibitors reduce desire and response [16]

Indirect

Reduction of self-image

Reduced by disfiguring surgeries, stomas, incontinence, altered appearance (e.g., drooling and altered faces of Parkinson’s, altered skin color and muscle wasting of renal failure)

Depressed mood

Depression and mood lability commonly accompany chronic illness; depression major determinant of sexual function in women with renal failure [17] or multiple sclerosis [18]; strong link between ED and subsequent depression [19]

Impaired mobility

Reduced ability to caress, hug, and hold a partner; to sexually self-stimulate, to stimulate a partner, to move into positions for intercourse, to pelvically thrust in spinal cord injury, Parkinson’s, brain injury, postamputation

Reduced energy

Fatigue may take its toll on sexuality especially desire, e.g., from renal failure or chemotherapy

Partnership difficulties

Difficulties finding a partner, dysfunction in the partner who assumes a care giver role, institutionalization, fear of becoming a burden to a partner, lack of independence. Relationship discord from stressors of living with medicalized lives (e.g., three times weekly hemodialysis)

Sense of loss of sexuality from imposed infertility

From surgery removing gonads or uterus, from chemotherapy or radiotherapy causing gonadal failure

Fear of sex worsening medical condition

Avoiding sex fearing a further stroke

From Basson et al. [5]

ED erectile dysfunction

The traditional sexual response cycle including excitement, plateau, orgasm, and resolution set the foundation for studying and categorizing sexual dysfunctions in men and women.

Sexual dysfunction generally is classified into four categories:

Desire disorders—lack of sexual desire or interest in sex: Sexual desire disorders or decreased libido (also known as hypoactive sexual desire disorder; HSDD) are characterized by reduction or absence for some period of time of sexual desire or libido for sexual activity or of sexual fantasies. The condition ranges from a general lack of sexual desire to a lack of sexual desire for the current partner, and it may have started after a period of normal sexual functioning or the person may always have had no/low sexual desire. Hypoactive sexual desire disorder has been reported in approximately 30 % of women and 15 % of men in population-based studies and is associated with a wide variety of medical and psychologic causes.
Arousal disorders—inability to become physically aroused or excited during sexual activity: Sexual arousal disorders, including erectile dysfunction in men and sexual arousal disorder in women, are found in 10–20 % (arousal and lubrication disorders are reported in 8–28 % of women) and are strongly age-related in men. They are characterized by a normal desire for sex but a difficulty or inability to become aroused or maintain arousal during sexual activity.
Orgasm disorders—delay or absence of orgasm (climax), ejaculation disorders (for male): Orgasmic disorder is relatively common in women, affecting about 10–25 % in community-based studies. In contrast, premature ejaculation is the most common sexual complaint of men, with a reporting rate of approximately 30 % in most studies. These conditions consist in a persistent or recurrent difficulty in achieving orgasm after sufficient sexual arousal and ongoing stimulation (anorgasmia or delayed ejaculation).

On the contrary, premature ejaculation (PE) occurs when a man experiences orgasm and expels semen soon after sexual activity and with minimal penile stimulation. However, there is no uniform cutoff defining “premature,” and values between 15 s and one minute are reported. Men’s typical ejaculatory latency is considered approximately 4–8 min [6].

Another uncommon ejaculation disorder is retrograde ejaculation, which occurs when semen enters the bladder instead of going out through the urethra during ejaculation. The main reason is an incomplete or ineffective closure of the bladder neck. It can be caused by medications, health conditions, or surgeries that affect the nerves or muscles that control the bladder opening.
Pain disorders—pain during intercourse: Sexual pain disorders have been reported in 10–15 % of women and less than 5 % of men. Dyspareunia (painful intercourse), vaginismus (an involuntary and painful spasm of the muscles of the vaginal wall that interferes with intercourse), and vulvodynia (burning vulvar pain not necessary related to sexual activity) are common sexual pain-related conditions. Dyspareunia may be caused by insufficient lubrication (vaginal dryness) in women. Poor lubrication may result from insufficient excitement and stimulation or from hormonal changes caused by menopause, pregnancy, or breast-feeding. It is unclear exactly what causes vaginismus, but it is thought that past sexual trauma (such as rape or abuse) may play a role.

However, there is considerable overlap between sexual dysfunctions, especially in women [7]. In patients with hypoactive sexual desire disorder, 41 % of women had at least one other sexual dysfunction and 18 % had diagnoses in all three categories (excluding pain disorders) [8].

17.2 Sexual Dysfunction and Pelvic Pain Disorders

Urinary voiding disorders have been strongly associated with sexual dysfunction. In women with urinary incontinence, 60 % had urinary incontinence during sexual intercourse, which significantly impacted their sex life [9]. Similarly, others have reported greater degrees of incontinence correlating with lower scores on the sexual function survey [10, 11].

The landmark National Health and Social Life Survey established a strong association between urinary tract symptoms and arousal disorders (odds ratio 4.2; 95 % confidence interval 2.75–5.89) and sexual pain disorders (odds ratio 7.61; 95 % confidence interval 4.06–14.26) [12]. The prevalence of sexual dysfunction secondary to sexual and pelvic pain ranges from 7 to 58 % [13].

In addition, women with genitourinary hypersensitivity disorders, classified by the International Continence Society as “genitourinary pain syndromes,” account for a large percentage of female patients who present to urogynecologic and sexual medicine practices [14].

Interstitial cystitis/painful bladder syndrome (IC/PBS) is a potential cause of SD and should be considered in the differential diagnosis for dyspareunia. Dyspareunia is estimated to occur in 49–90 % of women with IC/PBS, and urinary and pain symptoms are exacerbated following sexual activity [15].

Using the Female Sexual Distress Scale (FSDS), Peters et al. reported that IC/PBS patients had significantly more dyspareunia (74.6 % vs.29.9 %), more fear of pain (50.2 % vs. 13.5 %), and significantly less sexual desire and ability to achieve orgasm compared with healthy controls [16].

The prevalence of SD among women affected by laxity disorders of the pelvic floor is estimated to be 25–63 %. An estimated 11 % of these women undergo pelvic surgery for their laxity disorder, and the surgical repair procedure may either enhance or further impair their sexual function [12, 17].

Coital urinary incontinence (CUI) is another significant factor that affects sexual function and may be underdiagnosed in the clinical setting. Unless women are asked directly about its occurrence, they rarely report CUI spontaneously. Bachmann et al. noted that only 3 % of women self-reported sexual disturbances (including CUI); however, the prevalence increased to 20 % after direct questioning in an outpatient setting [18].

Surgical repair of pelvic organ prolapse or stress urinary incontinence, while generally beneficial for a woman’s quality of life, can also have negative effects [19]. Depending on the extent of surgical dissection, tissue damage, devascularization, and denervation involved, the result can be decreased vaginal blood flow and increased fibrosis, ultimately resulting in increased dyspareunia.

Sexual function improves in women following pelvic reconstructive surgery, but the improvement is more substantial following anterior repair either alone or in combination with a vaginal hysterectomy when compared with posterior repair [20].

For upper vaginal prolapse (uterine or vault), abdominal sacral colpopexy was associated with a lower rate of recurrent vault prolapse on examination and painful intercourse than with vaginal sacrospinous colpopexy [21].

The association between vaginal mesh and dyspareunia is well documented, especially after posterior repair, and the incidence of dyspareunia was increased associating levator myorrhaphy to posterior vaginal repair, with a de novo dyspareunia rate up to 20 % [22].

Despite the good functional results achieved by expert surgeons, large multicenter studies show that urogenital dysfunction remains a common problem after rectal cancer treatment. More than half of patients experience a deterioration in sexual function, consisting of ejaculatory problems and impotence in men and vaginal dryness and dyspareunia in women. Radiotherapy seems to have a role in the development of sexual dysfunction, without affecting urinary function [23].

Laparoscopic or robotic total mesorectal excision for rectal cancer is associated with significantly less deterioration in sexual function compared with open surgery. This effect is particularly pronounced in women [24].

Prostate cancer by itself reduces sexual desire and the frequency of sexual intercourse. Additionally, surgery or hormonal therapy to block testosterone further increases the frequency of erectile dysfunction. Erectile dysfunction following radical prostatectomy is primarily attributable to nerve injury caused by intraoperative nerve traction, thermal injury, ischemic injury, and local inflammatory reactions [25].

Women with fecal incontinence were as likely to engage in sexual relations as women without fecal incontinence; however, sexually active women with fecal incontinence had poorer sexual function and reported more dyspareunia, fear, and avoidance of sexual activity with greater partner problems than women without fecal incontinence [26].

17.3 General Treatment Principles

Treatment of sexual dysfunction is complicated by the lack of a single causative factor, limited proven treatment options, physician unfamiliarity with available treatments, overlap of different types of dysfunction, limited availability of treatment, and limited expertise in the treatment [27].

Most types of sexual dysfunction can be corrected by treating the underlying physical or psychological problems. Sexual therapy and education (e.g., cognitive behavior therapy, individual and couple therapy, physiotherapy) form the basis of treatment in most cases. The success of treatment for sexual dysfunction depends on the underlying cause of the problem and the outcome is good for dysfunction that is related to a treatable or reversible physical condition.

Substantial advances have occurred in the understanding of the pathophysiology of erectile dysfunction that led to the development of successful oral therapies, namely, the phosphodiesterase type 5 inhibitors, even if with limitations [28].

With the advent of phosphodiesterase type 5 inhibition as oral therapy, intracavernous injection of vasoactive agents has been relegated to second-line therapy for most patients with erectile dysfunction. However, the future of this category of agents remains attracting and an ever-expanding number and combination of agents are under investigation, making intracavernous injection more appealing as greater efficacy and tolerability and more rapid onset are attained [29].

Developments in the treatment of male erectile dysfunction have led to investigation of pharmacotherapy for the treatment of female sexual dysfunction, even if the benefit of hormonal and nonhormonal drugs is less demonstrated. Testosterone improves sexual function in postmenopausal women with hypoactive sexual desire disorder, although data on its long-term safety and effectiveness are lacking. Estrogen improves dyspareunia associated with vulvovaginal atrophy in postmenopausal women. Phosphodiesterase inhibitors have been shown to have limited benefit in small subsets of women with sexual dysfunction [30].

The myriad of therapeutic modalities for the treatment of female sexual dysfunction are summarized in Table 17.2, underlining the difficulties and the variability of the treatment of these diseases.

Table 17.2

Therapeutic modalities for the treatment of female sexual dysfunction

Behavioral therapy	Pharmacological therapy to bladder	Nonpharmacological therapy to pelvic floor	Therapy to vagina and vulva
Dietary modifications	Pentosan polysulfate sodium	Physical therapy	Lubricants
Stress reduction	Hydroxyzine hydrochloride	Myofascial massage	Estrogen creams
Relationship counseling and sex therapy	Amitriptyline hydrochloride	Thiele massage	Cromolyn
Lifestyle modifications	Cyclosporin A	Pharmacological therapy to pelvic floor	Capsaicin
Planned timing for intercourse	Intravesical bladder therapy/cocktails (dimethyl sulfoxide, heparin, lidocaine, Marcaine, steroids)	Muscle relaxants (Valium suppository)	Lidocaine gel
Pre- or postcoital bathing	Antispasmodics and antimuscarinics	Trigger point injections with local anesthetics and anti-inflammatory medications	Trigger point injections with local anesthetics and anti-inflammatory medications
Changing sexual positions	Botulinum toxin A	Botulinum toxin A
Noncoital sexual alternatives	Neuromodulation
Pre- and postcoital ice pack applied to genital and suprapublic areas	Medical: Gabapentin, Pregabalin
Vaginal diators	Surgical: Peripheral and sacral nerve stimulation

From Wehlbe et al. [31]

17.4 Electrotherapies for Sexual Dysfunctions

In this wide field, electrotherapies could have a potential role for treating specific symptoms in specific patients.

The principles of using electrical stimulation of peripheral nerves or nerve roots for restoring useful bladder, bowel, and sexual function after damage or disease of the central nervous system have been extensively reported in literature.

17.4.1 Neuromodulation for Female Sexual Dysfunction

Since the 1990s, sacral neuromodulation (SNM) is used in patients with refractory idiopathic symptoms of overactive bladder or nonobstructive urinary retention and then for fecal incontinence ad constipation. During routine follow-up for SNM, a number of patients spontaneously reported improved sexual functioning in comparison to before their SNM implantation.

The rationale of this result can be highlighted considering the common innervation of the involved structures through the pudendal and the hypogastric nerves.

However, even if the associations between lower urinary tract symptoms (LUTS) and female sexual dysfunction (FSD) are well documented, the question remains as to whether improved sexual function following SNM results from a reduction in LUTS and/or the associated improvement in quality of life (QoL) or if it occurs independently of changes in urinary function.

In the study of Gill et al. [32] in 33 patients affected by overactive bladder (OAB), a statistically significant improvement in urinary and sexual function occurred according to multiple metrics. Female Sexual Function Index (FSFI) and female sexual health questionnaire (FSHQ) were used to assess the sexual status and function of the patients, and also urinary function was assessed using validated instruments. Of the 33 patients only 10 where sexually active, and all the patients were in a menopausal status.

Validated urinary symptom and quality of life scores improved significantly. After treatment, most patients were incontinent less often with sexual activity and felt less restricted from sexual activity by fear of incontinence. Validated quantification of sexual function demonstrated significant improvements in overall sexual function, arousal, and satisfaction. In their results, improved sexual function was not significantly associated with improved urinary function after SNM despite apparent trends between the two.

Opposed to this result, studying a possible correlation between OAB symptoms improvement and sexual function improvement, Signorello et al. [33] analyzed the correlations between differences in Female Sexual Function Index (FSFI) scores and in clinical outcome and correlations between differences in FSFI, short form 36 (SF36), and incontinence quality of life index (IQoL) scores in 16 sexually active patients treated for OAB with SNM (15/16 were in menopause).

A significant correlation was found between clinical improvement and improvement in sexual function. No significant correlation was found between differences in FSFI and quality of life index (IQoL and SF36).

The authors suggest that improvement in the quality of sexual function in female patients with OAB correlates with improvement in urinary symptoms.

Pauls et al. [34] reported improvement in the FSFI score in 11 female patients with an SNM implant for LUTS. Significant improvement with regard to desire, lubrication, orgasm, satisfaction, and pain was found. No increase in score was noted for the arousal domain. Sexual arousal is a response to a sexually attractive stimulus and has both a physiological and a subjective component.

Interestingly, all patients in Pauls’ study improved on the FSFI, while only three subjects reported subjective improvement in sexual functioning.

Lombardi et al. [35] presented the results of their study on the effect of SNM on sexuality in 31 female patients. In this study, both patients with idiopathic and with neurological causes for their LUTS were included. Improvement in Female Sexual Distress Scale and FSFI after SNM was found in both patient groups.

Also Yih and colleagues [36] suggest a possible positive role of SNM, observing that sexual function improves along with urinary symptoms after neuromodulation in 167 patients with voiding symptoms evaluated with FSFI and Interstitial Cystitis Symptom-Problem Indices (ICSI-PI). Improved FSFI domains included desire, orgasm, satisfaction, and pain. Of the 74 sexually inactive patients at baseline, 10 became sexually active during the follow-up.

On the contrary Ingber et al. [37] used the FSFI on 54 female patients (27 sexually inactive) affected by OAB or painful bladder syndrome (PBS), who were scheduled to receive a neuromodulation implant. They administered the FSFI before and after 6 months from implantation, and they found no significant improvement in female sexual function.

Both neuromodulation by SNM and by pudendal implant techniques were used in their study.

Pudendal nerve stimulation for SD was also described in the paper of Peters [38], focusing on the technique and according to the original technique described by Spinelli [39].

Also the study of van Voskuilen et al. [40] on 8 patients treated with SNM for urgency symptoms (6 patients), urinary retention (1), and fecal incontinence (1) did not show a clear effect of SNM on sexual function, although there seems to be an improvement in orgasm scores. However, the lack of response on psychological questionnaires (5 questionnaires: Questionnaire for Screening for Sexual Dysfunctions, the Golombok-Rust Inventory of Sexual Satisfaction, the Symptom Checklist-90, the Maudsley Marital Questionnaire, and the McGill-Mah Orgasm Questionnaire) and the increase in vaginal pulse amplitude at plethysmography after SNM implantation could indicate that the improvement seems to be more physiologically than psychologically mediated.

Jarrett et al. [41] administered a self-written sexual questionnaire to patients with fecal incontinence before and after SNM implantation.

Nine of the 16 patients were sexually active. All nine patients reported that their sex life had been affected by fecal incontinence prior to SNM, and seven had felt benefit from implantation. The median improvement in their sex life was 40 %, and the percentage of improvement was inversely correlated to age.

Another study on the effects of SNM for LUTS on female sexual function was carried out by Zahibi et al. [42].

A characteristic of this study was that a large proportion of the patients had pelvic pain as well as LUTS. The study group chose to perform bilateral SNM and to position the electrodes epidurally in the sacral canal, thus stimulating the nerves of S2 and S4 as well as S3. A significant improvement in the FSFI was found. Results were better in patients who underwent the treatment for voiding dysfunction compared to those who had pain as their primary complaint. Domains with no significant improvement were desire and pain.

A concomitant improvement of LUTS and SD was noted even after percutaneous tibial nerve stimulation. Patients most likely to benefit were women, patients with an OAB, and subjective responders. The aspects of sexual life which mostly improved were overall satisfaction, libido, and frequency of sexual activities [43].

In front of these conflicting results, some factors need to be considered.

One of the most obvious causes for a not consistent concordance between the studies is the low number of patients tested. A reason for this might be that many patients with prolonged LUTS or fecal symptoms are not sexually active because of their complaints, and current sexual activity was not always an inclusion criterion. Considering sexually inactive patients in the measurement could be a selection bias.

Moreover, a large proportion of the group of patients who have urinary or fecal symptoms and are considered for SNM therapy are postmenopausal patients, as evident in the described papers.

Postmenopausal women report a relatively high rate of sexual dysfunction (higher than men). There is a marked decline in sexual interest and frequency of sexual activity. Lower estrogen levels after menopause may lead to changes in genital tissues and sexual responsiveness, and in this group considerable comorbidity can be found such as hysterectomy, vascular disease, diabetes, or arthrosis that might also preclude patients from engaging in sexual activity [44].

Previous pelvic surgery could be another influencing factor, as well as the lack of dedicated studies about SNM for SD, with SD as the main indication for the treatment.

There has been one case report on the use of sacral neuromodulation for the treatment of refractory vulvar vestibulitis syndrome.

Sacral neuromodulation was used, and at 6 weeks postoperatively, the patient reported that her pain had decreased from a 10/10 to a 2/10. At 24 months postimplantation, the patient continued to report decreased pain, was able to resume coitus, noted an increase in pain-free days, and a decrease in the intensity of pain during flares [45].

In another case report, a 51-year-old patient that presented with symptoms of lower urinary tract dysfunction and clitoral pain after an abdominal hysterectomy treated with SNM was described.

During test stimulation, she experienced only moderate improvement in voiding symptoms, but a striking improvement in clitoral pain symptoms. She underwent a two-stage implantation of a neurostimulator with a successful outcome after 6 months follow-up [46].

Govaert et al. [47] shows that SNM has an effect on uterus contractility. A decrease in frequency of contractions is seen with the SNM system switched on. However, as the measurements were performed in patients in the resting state, it is not clear what the effect of SNM on the uterus is while the patient has an orgasm.

Lastly, there are case reports on female patients who received lumbar epidural electrical stimulation, who have spontaneous orgasms when the simulator is turned on.

Through standard techniques, quadripolar or octopolar leads were placed in the epidural space percutaneously. The lead was maneuvered initially to an L1–L2 position and then repositioned based on feedback from the patient. The patients were allowed to utilize the device ad libitum for up to 9 days. These women described a greater frequency in sexual activity, increased lubrication, and overall satisfaction. A return of orgasmic capacity was found in 80 % (4/5) of patients having secondary anorgasmia with an average intensity of ≥3/5 while using the device. Once the device was removed, the patients returned to their previous anorgasmic status [48, 49].

17.4.2 Neuromodulation for Erectile Dysfunction

Anecdotally, most men report improved erectile functioning after neuromodulation for concomitant pelvic floor diseases. Research in monkeys, dogs, and rats has shown that electrical stimulation of the cavernous nerve results in an erection by causing increased arterial flow, relaxation of the cavernous muscles, and venous outflow restriction [50–52].

Shafik [53] implanted a cavernous nerve stimulation device in a series of 15 men for the treatment of erectile dysfunction. Cavernous nerve stimulation at a frequency of 10 Hz led to penile tumescence and an increase in intracavernous pressure but poor rigidity. When the stimulation frequency was increased to 60 Hz, penile tumescence and rigidity and intracavernous pressure increased, and full erection was achieved.

Additionally, Shafik’s study demonstrates that unilateral cavernous nerve stimulation is sufficient to induce erection.

Erectile dysfunction is a recognized, common adverse consequence of radical prostatectomy as well as various other pelvic surgeries. While a host of management options have been considered to decrease this complication, neuromodulatory therapy has recently been advanced as an intervention that may be applied for this purpose. Neuromodulatory therapy offers a therapeutic approach for addressing the neuropathic changes of the penis that occurs in this context with the goal of maximally preserving erectile function postoperatively [54].

In 16 men undergoing retropubic radical prostatectomy and in 6 undergoing penile surgery for venous leakage, Lue et al. [55] applied electrical stimulation to the prostatic apex bilaterally (prostatectomy group) or to the hilum of the penis (venous surgery group). Electrical stimulation produced visible erection in 8 of the 16 prostatectomy patients and an increase in intracavernous pressure in 5 of the 6 venous surgery patients.

Burnett et al. [56] explored the feasibility of using an implantable electrode array for cavernous nerve stimulation for patients undergoing nerve-sparing prostatectomy. The implantable electrode array was placed over the neurovascular bundles (20 Hz frequency, 260 μs pulse width, 5–60 mA amplitude up to 10 min) in 12 patients undergoing open retropubic radical prostatectomy, and penile circumference increases were measured. Six of 12 (50 %) patients demonstrated a significant increase in penile circumference after stimulation.

To determine if intraoperative stimulation of the cavernous nerves while monitoring changes in penile tumescence could be useful to map the course of these nerves and would result in an improvement in nerve sparing and erectile function after radical prostatectomy, Klotz and Herschorn [57] studied 23 patients.

A cavernous nerve stimulator and tumescence-monitoring device was used during radical prostatectomy to identify the course of the cavernous nerves and guide the surgeon in avoiding nerve damage. Nineteen of 21 patients reported erectile function preoperatively. Seventeen (89 %) of 19 patients demonstrated a tumescence response during surgery. Sixteen (94 %) of the 17 patients who demonstrated a response to nerve stimulation and for whom the surgery was guided by the tumescence response reported the ability to have erections after surgery.

The same author confirmed the results in a multicentric prospective, randomized, single-blinded study, performed on 61 patients at 6 centers [58]. Patients had elected to undergo nerve-sparing prostatectomy and had normal preoperative erectile function documented by the Sexual Function Inventory Questionnaire (SFIQ) and RigiScan parallel testing. At 1 year, there was substantial improvement in erectile function in the group in which the procedure was performed assisted by the neurostimulation mapping. This group had a mean of 15.9 min of greater than 60 % nocturnal tumescence compared to 2.1 min in the conventional nerve sparing group (p < 0.024).

Only gold members can continue reading. Log In or Register to continue