Pleural effusion following use of saline and fluid anti-adhesion agents at laparoscopic surgery—a case series of three patients


Rashmi Ronghe, Specialist Registrar in Obstetrics and Gynaecology, Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, UK. Email

Case report

Laparoscopy is widely used for diagnostic as well as therapeutic purposes in gynaecology. The green top guidelines on the management of endometriosis produced by the Royal College of Obstetricians and Gynaecologists recommend advanced laparoscopic surgery for treatment of severe endometriosis. Here, we present three very unusual cases of hydrothorax developed as a complication of advanced laparoscopic surgery for endometriosis.

The first patient was a 38-year-old nulliparous woman with BMI of 25.7 with known severe endometriosis referred to gynaecology clinic for consideration of advanced laparoscopic surgery. The patient had undergone diagnostic laparoscopy with no complications in the past. She was fit with no significant past medical history. Preoperative investigations were unremarkable.

The patient was intubated, paralysed and ventilated during the operation. She was in lithotomy position with modest Trendlenburg tilt. An average intraperitoneal pressure of 15 mmHg was maintained during the procedure.

Introperatively, she was found to have stage IV endometriosis with dense adhesions involving both tubes and ovaries. Laparoscopic adhesiolysis and laser treatment of endometriosis was carried out. The procedure lasted for 1 hour. A total of 3 l of normal saline were used for suction irrigation. Although the exact irrigation balance was not recorded, it was assumed that the majority of the fluid was recovered. At the end of the procedure, 1.5 l of ADEPT (4% icodextrin) was left in the peritoneal cavity.

About 40–45 minutes into the operation, it became difficult to ventilate the patient. Her airway pressure increased, end tidal CO2 increased and oxygen saturation dropped from 100 to 90% on an FiO2 of 0.5. This occurred over a period of 5–10 minutes. Saturations improved to 96% on 100% oxygen. At this stage, almost 2500 ml of saline had been used for irrigation. On auscultation, breath sounds were present bilaterally. The operation was completed in the next 20 minutes. The patient required to be ventilated with 100% oxygen to maintain saturations over 90%. It was deemed unsafe to extubate the patient. Arterial blood gas analysis showed Po2 of 11.2 kPa and Pco2 of 7.8 kPa on 100% oxygen with 5 cm positive end-expiratory pressure (PEEP).

At this stage, breath sounds were noticed to be diminished over the right side of the chest. A portable chest X-ray showed a white out of the right side (Figure 1). An ultrasound scan was performed, which demonstrated the presence of a large amount of fluid in the right hemithorax. A chest drain was inserted in the right pleural cavity and immediately, almost a litre of fluid was drained. Repeat arterial blood gas analysis showed Po2 of 43.2 kPa and Pco2 of 5.3 kPa.

Figure 1.

 Right Hydrothorax.

The right lung did not inflate adequately and thus the patient was transferred to ITU. She was ventilated overnight and extubated the following morning and recovered rapidly. The patient was discharged home on the subsequent day from the ITU. Cytology and microbiology examination of the fluid was negative.

The patient was reviewed for follow-up visit in the gynaecology clinic 3 weeks postoperatively and was well.

The second patient was a 38-year-old nulliparous woman with the BMI of 24 and known Stage IV endometriosis. She had undergone laparoscopy and laser treatment of endometriosis 3 years ago without any complications. Normal saline was used for irrigation and 1 l of ADEPT was left in the peritoneal cavity at the end of the procedure. She then had two cycles of IVF with unsuccessful outcomes and was referred from the assisted conception unit for further laparoscopic surgery in view of worsening symptoms.

The patient was intubated, paralysed and ventilated for the operation and the average intraperitoneal pressure of 15 mmHg was maintained.

Intra-operatively, she was found to have stage IV endometriosis with adhesions involving left tube and ovary. Endometriosis was present on both uterosacral ligaments. The uterus was retroverted and fixed. Laparoscopic laser treatment and adhesiolysis was performed. The procedure lasted for 80 minutes. The exact amount of saline used for the procedure and the irrigation balance was not recorded. One and a half litres of ADEPT was left in the peritoneal cavity at the end of the procedure.

The patient was well intra-operatively with SpO2 of 98–100% and there were no problems in the recovery room. She started complaining of difficulty in breathing and right-sided shoulder-tip pain on the night of the operation. At 9 p.m. in the evening after the operation, her oxygen saturation dropped to 95% on air; her respiratory rate was 30/minute and pulse was 100 bpm. Chest X-ray demonstrated a left-sided basal effusion. Arterial blood gas analysis showed Pco2 of 5.4 kPa and Po2 of 12.4 kPa on 60% oxygen. Needle aspiration of the fluid was performed and sent for cytology and biochemistry. She was transferred to HDU for observation. Her SpO2 improved with oxygen. She recovered well and was discharged home on the third postoperative day.

The patient was reviewed in the gynaecology clinic 5 weeks after the operation and was well.

The third patient was a 44-year-old nulliparous woman with a long standing history of endometriosis. She had a BMI of 23.8. She had undergone laparoscopic laser treatment of endometriosis twice in the past 5 years. Both these times, the procedures had lasted for 20 and 40 minutes respectively. Unfortunately, there is no record of amount of fluid used for irrigation or the amount of ADEPT left in the peritoneal cavity at the end of these procedures.

The patient was intubated, paralysed and ventilated for the operation. She was found to have widespread endometriosis, which was treated with laser. Normal saline was used for suction irrigation (balance not recorded) and 1.5 l of ADEPT was left in the peritoneal cavity at the end of the procedure. The procedure lasted for 30 minutes.

The oxygen saturations were unchanged throughout the procedure and she was well in the immediate postoperative period. Her SpO2 dropped to 95% on the evening of the operation. By 6 a.m., the following morning, the SpO2 dropped to 93% on air and she was started on 2 l of oxygen. She started complaining of discomfort while breathing and right-sided shoulder-tip pain. A chest X-ray reported, ‘Right-sided pleural effusion extending into right mid zone along oblique fissure with right basal lung collapse’. The patient was transferred to respiratory ward. Repeat chest X-ray in 48 hours showed that the effusion had resolved spontaneously.

The patient was followed up in gynaecology clinic in 3 months and was well.


Development of hydrothorax during laparoscopic surgery in gynaecology is rare and hence is not discussed as one of the potential complications when counselling patients for the procedure. We describe this rare complication following laparoscopic laser treatment of severe endometriosis. It is likely that the hydrothorax developed as a consequence of the movement of fluid from peritoneal cavity into pleural space through small congenital defects in diaphragm. In the first case, the SpO2 dropped from 98 to 90% about 40 minutes after the start of surgery. By this stage, about 2500 l of saline was used for irrigation. Although it was assumed that most of it was aspirated, some must have passed to the upper abdominal cavity. This was a particularly difficult case, which lasted for 60 minutes. Unfortunately, the exact irrigation balance was not recorded for any of these cases. In the second patient as well, the surgery was difficult and had lasted for 80 minutes. Interestingly, the second patient had undergone a similar procedure 3 years previously without any complications. One litre of ADEPT was left in the peritoneal cavity at the end of the procedure at that time and the procedure had lasted for a shorter time. The third patient had also undergone similar procedure twice before without any complications.

The first two cases were carried out on the same operating list. The third patient who had undergone her surgery 3 months prior to the first two cases, however, was identified to have this complication retrospectively. All three cases were carried out by the same gynaecologist with 17 years of experience of advanced laparoscopic surgery. There was no significant intra-operative bleeding in any of these cases. As two of the cases occurred on the same list, the theatre insufflation instrument was checked to ensure that the intra-peritoneal pressure was indeed maintained at 15 mmHg during the procedures. In the last two cases, the hydrothorax completely resolved spontaneously.

Other studies have reported channels of communication between the pleural and peritoneal cavities.1LeVeen et al.2 reported cases of development of hydrothorax in patients with ascites. They demonstrated small defects in diaphragm covered by pleuroperitoneum, which ruptured with increased abdominal pressure. The majority of these cases had right-sided hydrothorax. Similarly, Strauss and Boyer3 reported cases of hepatic hydrothorax involving similar mechanism. Hepatic hydrothorax is defined as the presence of a pleural effusion in a patient with liver cirrhosis, in the absence of cardiopulmonary disease. Hydrothorax complicating peritoneal dialysis is also well reported.4 The prevalence of hydrothorax in cases of liver cirrhosis and peritoneal dialysis is 6 and 2% respectively.6 Over the period of 3 months, when these three cases underwent their surgery, a total of 49 cases of operative laparoscopy were carried out in our unit. There have been no further new cases with this complication in our unit since then (12 months).

ADEPT is a 4% solution of icodextrin, an α-1, 4 linked glucose polymer. It is a well-established agent, which prevents adhesion formation after gynaecological surgery. Icodextrin is retained in the peritoneal cavity for long periods of time because it is metabolised by amylase. As there is no amylase present within human peritoneal cavity, the fluid is cleared by lymphatic drainage. The safety profile of ADEPT in gynaecological surgery has been proven with clinical trials such as PAMELA study and through ARIEL registry. However, there are rare reports of pulmonary oedema and pleural effusion from clinical experience with ADEPT. Doumplis et al.5 reported two cases of symptomatic pleural effusion after a laparoscopic and a laparotomy surgery when icodextrin was used.

Although this complication is rare, it has been reported in the literature.6 The first patient in our series developed hydrothorax prior to the infusion of ADEPT, that is, it was saline. The possible factors contributing to the development of hydrothorax in these patients could be the prolonged duration of procedure, amount of fluid in the peritoneal cavity, Trendlenburg position and increased intra-abdominal pressure. We believe that surgeons and anaesthetists should be aware of hydrothorax as a potential complication of advanced laparoscopic surgery. Irrigation balance should be monitored closely and anaesthetists and surgeons should be notified of negative balance. It is important to level the patient to allow irrigation fluid to drain back in the peritoneal cavity and to be retrieved particularly in case of ventilation difficulties. ADEPT comes in 1.5 l bag, but only 1 l should be installed in the peritoneal cavity. The rest can be used for irrigation.

Disclosure of interest

Not applicable.

Contribution to authorship

All three authors have made substantial contributions to the conception and design of the case report. The original draft of the article was prepared by Dr Ronghe, which was then revised critically by Dr Bjornsson and Dr Hannah for important intellectual content. All the authors then approved the final version of the article to be published.

Details of ethics approval

Ethics approval was not required for these case reports.


Funding was not required for theses case reports.


We would like to thank the three patients involved in these case reports who kindly gave us written permission to present their cases.