Ann Thorac Surg 2004;77:1016-1021
© 2004 The Society of Thoracic Surgeons
Original article: general thoracic
Pleural perfusion thermo-chemotherapy under VATS: a new less invasive modality for advanced lung cancer with pleural spread
Norihisa Shigemura, MDa*,
Akinori Akashi, MDa,
Tomoyuki Nakagiri, MDa,
Kenji Hazama, MDb,
Mitsunori Ohta, MDb,
Hikaru Matsuda, MDb
a Division of General Thoracic Surgery, Takarazuka Municipal Hospital, Hyogo, Japan
b Department of Surgery, Osaka University Graduate School of Medicine, Osaka, Japan
Accepted for publication August 19, 2003.
* Address reprint requests to Dr Shigemura, Department of Surgery, Osaka University Graduate School of Medicine, E-1, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan
e-mail: n-shige{at}blue.ocn.ne.jp
 |
Abstract
|
|---|
BACKGROUND: We conducted a trial of a new less invasive, locoregional modality for lung cancer with pleural spread. This study was planned to investigate the feasibility, safety, and pharmacokinetics of pleural perfusion thermochemotherapy (PPTC) under video-assisted thoracoscopic surgery (VATS) and its modified method with a short perfusion time for preventing heat damage to the lung during the procedure.
METHODS: Seventeen patients, 59 to 79 years old, underwent surgical resection of the primary lesions and PPTC under VATS without thoracotomy. All had pleural spread with malignant effusion due to lung cancer proven before the treatment. PPTC consists of irrigating the pleural space with 42°C saline solution containing cisplatin (200 mg/m2) using a devised circuit. The time for perfusion was two hours in 10 patients (group L), and one hour in 7 patients (group S).
RESULTS: All patients successfully completed this treatment with acceptable toxicities. The pharmacokinetic analysis showed that high platinum levels for the regional pleural exposure, which was 20- to 40-fold greater than those for the plasma in both groups. These values were equivalent between the groups, although the levels for the plasma were higher in group L than in group S. Postoperative lung damage was seen in 4 patients with no serious conditions in group L, and none in group S. The median survival for the L and S groups was 17 and 19 months, respectively.
CONCLUSIONS: This less invasive modality seems to offer a safe, feasible, and pharmacokinetically advantageous procedure to have excellent local control for lung cancer with pleural spread.
|
Introduction
|
|---|
Malignant pleural effusion is a common complication of advanced lung cancer. It is often associated with damaged quality of life, significant morbidity, and a short life expectancy. While many different treatment modalities, the main of which has been systemic chemotherapy, have been employed for lung cancer with carcinomatous pleuritis, no standard treatment has been established and therapeutic outcomes remain poor [1–3]. Recent reports have demonstrated the effectiveness of pleural perfusion thermo-chemotherapy (PPTC) [4, 5], which consists of synergistic coadministration of cis-diamminedichloroplatinum (CDDP) and thermotherapy. This treatment has been performed mainly using a method to allow direct irrigation of the thoracic cavity with thoracotomy [5]. Although PPTC has been employed in several institutions and reported relatively favorable results, pulmonary damage may occur secondary to the direct action of heat. For this reason, this method occasionally may not be appropriate for use in borderline patients with limited pulmonary function and cardiac complications, or in malnourished elderly patients, because the operative burden including thoracotomy is more invasive for them [4, 5].
To overcome these problems and to develop the indications for PPTC, we adapted PPTC for the video-assisted thoracoscopic surgery (VATS) approach, for which we have expected equivalent results to conventional PPTC in a less invasive and safe manner. In the present study, we describe early and midterm results of the use of PPTC under VATS for the treatment of lung cancer complicated by carcinomatous pleuritis, and discuss the effects of perfusion time on resulting heat damage to the lung.
|
Patients and methods
|
|---|
During January 2000 through April 2002, PPTC under VATS was carried out in seventeen patients with lung cancer complicated by malignant pleural effusion. Patients were included in the study if they had nonsmall cell lung carcinoma with proven pleural spread, without extrathoracic metastases, and a cardiorespiratory reserve allowing the required resection. Exclusion criteria included: (1) severe cardiac complications, (2) creatinine clearance less than 60 mL/min, (3) white blood cell count less than 4,000/µL, platelet count less than 75,000/µL, serum bilirubin and serum glutamic-oxaloacetic transaminase more than 2x normal, and (4) a history of thoracotomy or instillation of drugs in the operated side. The study was approved by the Institutional Review Board of Takarazuka Municipal Hospital for a phase I study. Additionally, the potential risks of treatment, standard and alternative treatment modalities, were explained to each patient and written, informed consent was obtained before treatment. Preoperative staging studies to exclude extrathoracic metastasis included a computed tomographic (CT) scan of the chest and abdomen, magnetic resonance imaging of the brain, and general bone scintigraphy.
There were 11 male and 6 female patients, and the ages ranged from 59 to 79 years (mean, 69.9 years). Pleural effusion was present by chest roentgenogram studies or computed tomography in all patients at the time of admission, and pleurocentesis was performed before PPTC under VATS. The results of cytologic examination confirmed class V lung cancer and carcinomatous pleuritis in all patients. The degree of dissemination detected at the time of the operation was classified as follows according to the criteria of the Japan Lung Cancer Society: D1, less than 10 visible dissemination nodules; D2, more than 11 visible disseminated nodules; D0, no visible nodules but cytology-positive effusion.
Perfusion technique
Pleural perfusion thermo-chemotherapy under VATS was performed generally according to the conventional method [5], with all procedures carried out under thoracoscopic guidance without thoracotomy. Using one-lung ventilation and general anesthesia, each patient was placed in the lateral position. The following ports were established: a 10-mm port was placed in the fifth intercostal space along the anterior axillary line (port 1); a 10-mm port was placed in the sixth intercostal space along the posterior axillary line (port 2); and a 5-mm port was placed in the third intercostal space along the anterior axillary line (port 3). A thoracoscope was inserted through port 1, while the remaining ports were utilized to perform lobectomy or partial resection under VATS, thereby excising tumor tissue and obtaining material for a pleural biopsy. An intrapleural temperature probe was subsequently inserted and placed under the intercostal pleura via port 3. Next, an irrigation inlet tube (10 mm) was inserted through port 1, and an irrigation outlet tube (10 mm) was inserted through port 2. These tubes were connected to a standard extracorporal circuit (heat exchanger, roller pump, and reservoir: CP FOUR BCP, MERA, Tokyo, Japan) as shown in Figure 1, and the circuits were primed with a mean volume of 3 L of saline solution; the pleural space perfusion flow was maintained between 800 and 1000 mL/min, as monitored with an electromagnetic flow probe. Because a temperature of approximately 45°C on the heat exchanger was selected, the minimal and maximal pleural space perfusion temperatures were maintained between 42°C and 42.5°C by the intrapleural temperature. After confirming that a temperature of 42°C was achieved by the intrapleural temperature probe, 200 mg/m2 of CDDP was administered; this corresponded to the beginning of the experimental perfusion period. The total dose of CDDP was: 250 mg in 3 patients, 270 mg in 6, 300 mg in 5, 320 mg in 2, and 340 mg in 1 patient. To prevent postoperative pulmonary edema, 1000 mg of methylprednisolone was administered intravenously during PPTC. Once perfusion was completed (for one or two hours), the solution in the thoracic cavity was removed as completely as possible, and a drainage tube was placed in the thoracic cavity.
View larger version (22K):
[in this window]
[in a new window]
|
Fig 1. Operative scheme of thoracoscopic (PPTC under VATS). Two tubes for irrigation inlet and outlet, which were inserted into the pleural cavity under thoracoscopic guidance without thoracotomy, were connected to a standard extracorporeal circuit (heat exchanger, roller pump, and reservoir). (CDDP = cis-diamminedichloroplatinum; PPTC = pleural perfusion thermo-chemotherapy; VATS = video-assisted thoracoscopic surgery.)
|
|
Pharmacokinetic analysis of CDDP in serum and effusion
Concentrations of CDDP were measured by collection of pleural effusion and blood samples before PPTC and at 1, 2, 24, and 48 hours after the start of perfusion. Each sample was centrifuged at 1000 rpm for 15 minutes, and the levels of total platinum (free platinum + protein-bound platinum) and free platinum were measured according to the method as previously reported in a study by Ratto and colleagues [6]. Animal studies [7] have demonstrated that the tumoricidal activity of CDDP is dependent on the concentration of free platinum. Therefore, we adopted the method of measuring mentioned above. Additionally, the area under the concentration (AUC) multiplied by the time curve from 0 to 48 hours was estimated by the linear trapezoidal rule. The pharmacokinetic advantage derived from intrapleural drug administration was determined by dividing the regional (pleural space perfusate) AUC by the systemic (plasma) AUC. Peripheral blood, biochemical, and blood gas analyses were also performed. A Student's t test was used for statistical analysis, and a p value less than 0.05 was considered significant.
Effects of perfusion time (one hour or two hours) in PPTC under VATS
In all 17 patients, PPTC was performed following resection of the primary lesion. To assess the effect of perfusion time on heat damage to the lung in PPTC under VATS, the results were compared between those with perfusion time for two hours and those for an hour. A longer two-hour perfusion was utilized in ten patients (L group), and a one-hour perfusion time was utilized in the remaining seven patients (S group). Table 1 summarizes patient demographics, clinical diagnosis, and tumor resection techniques for both groups. The concentration of CDDP, therapeutic results, postoperative complications (toxicities) including pulmonary edema, and their prognosis were compared between the two groups. Toxicity was graded according to the National Cancer Institute Common Toxicity Criteria. Pulmonary edema that was confirmed by rales over both lungs and signs of pulmonary congestion on chest radiographs after the operation was evaluated at the same time by the additional data of noninvasive blood pressure (BP), pulse oximetric saturation (Spo2), heart rate, respiratory rate, and arterial blood gas levels recorded at baseline (on room air).
Follow-up examinations were done every second month. A CT scan of the chest was performed every third month, and a CT scan of the abdomen and magnetic resonance imaging of the brain were performed every sixth month. When it was necessary to differentiate between pleural scarring and tumor recurrence, magnetic resonance imaging of the chest was added to a CT scan. The follow-up was updated on March 31, 2003.
|
Results
|
|---|
The pleural findings at the time of the operation showed five patients in the D1 category, four patients in the D2, and one patient in the D0 category from the L group. On the other hand, from the S group, four patients in the D1, and three in the D2 category were shown. Treatment was completed in all patients as expected.
Figure 2 shows the changes of platinum concentration in pleural effusion. In the L group, total platinum decreased to 45 µg/mL at one hour after the start of perfusion and gradually decreased to 40 µg/mL at two hours after the start of perfusion. The level of total platinum at 12, 24, and 48 hours after the start of perfusion was 20, 10, and 8 µg/mL, respectively. The level of free platinum at 1, 2, and 48 hours after the start of perfusion was 36, 30, and 2 µg/mL, respectively. For the S group, the level of total platinum at 1, 2, 12, and 24 hours after the start of perfusion was 38, 35, 18, and 8 µg/mL, respectively, and free platinum was 33, 28, 12, and 6 µg/mL, respectively. When compared to the L group, the level of total platinum in the S group was lower at 1, 2, 24, and 48 hours after the start of perfusion, while the level of free platinum for the S group was higher at 24 and 48 hours after the start of perfusion. Nonetheless, there was no significant difference in these parameters when comparing the two groups.
View larger version (14K):
[in this window]
[in a new window]
|
Fig 2. Changes of concentration of total and free platinum in pleural effusion of patients with carcinomatous pleuritis following PPTC with CDDP 200 mg/m2 for two-hour perfusion (group L) and one-hour perfusion (group S) methods, respectively. (CDDP = cis-diamminedichloroplatinum; PPTC = pleural perfusion thermo-chemotherapy.)
|
|
Figure 3 shows the changes in serum platinum concentrations. The level of total platinum peaked at 2.10 µg/mL at two hours after the start of perfusion in the L group and at 2.00 µg/mL at one hour after the start of perfusion in the S group. While the peak concentration of total platinum was slightly lower for the S group, there was no significant difference between the two groups. At 24 hours after the start of perfusion, the mean total platinum concentration for the L and S groups was 0.96 and 0.75 µg/mL, respectively. The level of free platinum at 1, 2, and 24 hours after the start of perfusion was 1.0, 0.80, and 0.50 µg/mL, respectively, in the L group and was 0.8, 0.6, and 0.3 µg/mL, respectively, in the S group. While the level of free platinum was greater than 1 µg/mL at all time points in the L group, it was never above 1 µg/mL at any time points in the S group.
View larger version (16K):
[in this window]
[in a new window]
|
Fig 3. Changes of concentration of total and free platinum in serum following PPTC with CDDP 200 mg/m2 for two-hour perfusion (group L) and one-hour perfusion (group S) methods, respectively. (CDDP = cis-diamminedichloroplatinum; PPTC = pleural perfusion thermo-chemotherapy.)
|
|
The data shown in Table 2 on total systemic and regional (pleural space perfusate) exposure to platinum during perfusion (AUC, 0 to 48 hours) and their mean ratio (AUC regional/AUC systemic values) revealed that systemic absorption in free cisplatin from the circuit was greater in group L than in group S, whereas the regional values were similar (736 ± 54 and 755 ± 69, respectively). The mean AUC regional (AUC systemic) values were 23 ± 9.2 in group L, and 37 ± 7.7 in group S.
Postoperative cytologic examination of pleural effusion for tumor cells was negative in 7 of 10 patients from the L group (70%) and in 5 of 7 patients from the S group (71%). Although no life-threatening postoperative complications occurred, radiographic evidence of mild pulmonary edema was present in 4 patients from the L group. Their baseline hemodynamic and respiratory parameters, the data for which were collected before therapy, are reported in Table 3. All four patients improved after conventional medical treatment with oxygen supply, diuretics, and vasoactive drugs. Ventilator assistance was not necessary for them, although a 77-year old woman with chronic pulmonary disease showed the lowest Spo2 (85%) with stable circulatory conditions. The other major toxicities are shown in Table 4. Three patients in the L group and one in the S group experienced grade 1 to 2 nausea and vomiting. Grade 1 to 2 renal dysfunction occurred in two from the L group and one from the S group. The maximum postoperative serum creatinine of these patients was 2.5 mg/dL, which was normalized after adequate hydration treatment. One patient manifested grade 2 cardiopulmonary symptoms (sinus bradycardia) temporally, and we did not resolve the cause. Furthermore, two patients from the L group had a persistent air leak for 10 days, which resolved on its own.
The follow-up ranged between 8 and 36 months (median, 28 months). Thirteen patients are alive 13 to 36 months after operation. All patients that survived were free of disease, except two who had a suspected contralateral pleural recurrence. There was no evidence of ipsilateral pulmonary or pleural disease in all survivors. Four patients died 8 to 19 months after operation. Three of them (all N2) died from systemic disease progression, and one from aspiration pneumonia aggravated. The one-year overall survival rates for the entire group, for the L group, and for the S group were 78%, 75%, and 80%, respectively. The median survival times were 17 for the L group and 19 months for the S group.
|
Comment
|
|---|
Pleural perfusion thermo-chemotherapy has recently been utilized as an efficacious treatment for intrathoracic malignancies such as invasive thymoma, thymic cancer, malignant mesothelioma, or advanced lung cancer [6, 8, 9]. Especially for advanced lung cancer with carcinomatous pleuritis, PPTC is already employed by several institutions, and in vitro studies have demonstrated that the combination of thermotherapy and CDDP administration is markedly effective in killing cancer cells [10]. However, in past reports PPTC was indicated only in patients with carcinomatous pleuritis discovered first at the time of thoracotomy, and not for the more advanced cases with major malignant effusion. Additionally, PPTC utilizing thoracotomy is invasive and is therefore considered to be inappropriate for the frail, borderline patients with limited pulmonary function, cardiac complication, or in malnourished elderly patients [4, 5]. Pleural perfusion thermo-chemotherapy is a promising treatment that is expected to have complete local control for localized advanced malignant cancer and may be useful in the treatment of more advanced carcinomatous pleuritis accompanied by major pleural effusion. Therefore, we consider that performing PPTC under the VATS approach reduces the degree of surgical invasion and potentially expands the range of indications for which PPTC may be adapted. Furthermore, when compared to conventional PPTC using thoracotomy, PPTC under VATS can be performed in the completely closed circuit without opening the chest, and as a result has the advantage that stable perfusion temperature and CDDP concentration can be maintained easily and quickly.
Thermotherapy can be divided into electromagnetic wave-based therapy and perfusion-based therapy using a modified extracorporeal circuit [4, 5, 8]. With the latter method (PPTC), heat conduction of the pleura can be achieved regardless of respiration, but the toxicities, including the effects of directly exposing the lung to heat, would not be insignificant. In malignant pleural mesothelioma and thymic malignancies that are characteristic of their aggressive local behavior, the use of PPTC has been reported and studied occasionally; however, in carcinomatous pleuritis due to lung cancer, the adequate, safe dose of antineoplastic drug (cisplatin) and thermal dose have not been decided definitely so far. Therefore, we tried in this study to determine them with acceptable toxicity.
On the issue of the dose of cisplatin, Zimm and colleagues reported in their phaseIstudy that the highest dose of cisplatin that can be administered safely was clearly 200 mg/m2 [11], and several reports confirmed its effectiveness and safety in PPTC for lung cancer [5, 9]. Based on these findings, we selected a 200 mg/m2 dose of cisplatin for intrapleural perfusion. For the thermal dose, which is the product of the temperature differential and the time interval at elevated temperature, the time interval has not been defined, and varied from one to two hours. In the previous reports, the majority adopted the two-hour method as Matsuzaki and colleagues did in their study with the thoracotomy method [5], and we also treated the first patients for two hours. However, they reported with a caution that radiologic evidence of pulmonary edema was present in 2 of 12 patients (16%) postoperatively. While intraoperative intravenous steroid administration was reported to be effective for the prevention of pulmonary edema [7], steroid administration may attenuate the tumoricidal activity of anticancer agents. In the current study, mild postoperative pulmonary edema was seen in 4 of 17 patients (23%) despite administration of steroidal drugs after the start of PPTC under VATS.
To overcome this serious problem of heat damage to the lung in PPTC, we have performed the trials of shortening the perfusion time while maintaining the equivalent efficacy in PPTC under VATS. As a result of pharmacokinetic analysis of CDDP, there were no significant differences in total and free platinum concentration in pleural effusion between the L (two-hour) and S (one-hour) groups. Although a decrease in the level of platinum secondary to increased absorption of platinum by the parietal pleura was found in the early stage of perfusion (eg, early rapid drop), free platinum concentrations, which were an active form and had cytotoxic activity, were maintained in the one-hour method up to 48 hours after the start at levels high enough for treatment of pleural dissemination [9]. In addition, our pharmacokinetic analysis on the AUC also showed the equivalent effectiveness in the short perfusion method to that for two hours, and the highly pharmacological advantage for intrapleural perfusion with free platinum in both groups (AUC regional/plasma ratio was as high as 20- to 40-fold). These data are in agreement with those reported by Ratto and colleagues in the malignant mesothelioma cases, indicating the important theoretic advantages for PPTC under VATS [6]. However, as they also mentioned in their article, this pharmacokinetic advantage for cisplatin pleural exposure does not necessarily translate into an enhanced delivery of the drug to target tissues. Therefore, whether a high delivery of the drug to the tumor results in prolonged disease-free and overall survival remains to be determined completely.
The level of free platinum concentration in serum was below 1 µg/mL in the S group at 1, 2, and 24 hours after the start of perfusion, which was significantly lower when compared to the L group. The AUC plasma values in group S were also significantly lower than in group L. The low frequencies of postoperative side effects such as nephrotoxicity and digestive tract toxicity for the S group should be ascribed to this fact.
With regards to the toxicity associated with heat damage of the lung, pulmonary edema was not seen in any of the S group patients. Heat-induced organ damage is irreversible, and especially heat-induced damage of the lung can be lethal as Rusch and colleagues indicated [12]. In our cases, all four patients with the disease fortunately responded to conventional medical treatment without ventilator assistance. However, as shown in Table 3, this critical toxicity tended to be frequent in elderly patients (over seventy) in our study, which would require further careful consideration. Pulmonary edema, one of the most important complications associated with thermo-chemotherapy, is a warning sign of severe and irreversible organ injury. On this issue, the present study provides the solution that the short perfusion method (modified PPTC under VATS) has comparable therapeutic efficiency to the conventional method while simultaneously minimizing heat-induced lung damage. At the same time, the results of our trial may suggest the advantages over the conventional, open PPTC method in the safety and indications, since the more elderly patients were included in our study (mean age, 69.9 years).
In summary, the present study has several limitations. There was no control group, or prospective design in phase I trial to our study, which utilized a relatively small number of patients and a limited follow-up period. However, the results do suggest that PPTC under VATS is a safe, facile, and effective procedure with a high local control for lung cancer complicated by carcinomatous pleuritis. Furthermore, PPTC under VATS with a short perfusion method was just as effective as the conventional method, may result in reduced lung damage due to heat, and be adapted for more frail, elderly, and borderline patients with multiple comorbidities, realizing a new less invasive modality for intrathoracic malignancies.
|
References
|
|---|
- Martini N., Bains M.S., Beattie E.J. Indications for pleurectomy in malignant effusion. Cancer 1975;35:734-738.[Medline]
- Ruckdeschel J.C. Management of malignant pleural effusion: an overview. Semin Oncol 1988;15:24-28.[Medline]
- Bunn P.A., Jr The role of systemic chemotherapy in non-small cell lung cancer. In: Bunn P.A., Jr, ed. Current Topics in Lung Cancer. Berlin: Springer-Verlag, 1991:33-46.
- Kodama K., Doi O., Higashiyama M., et al. Long-term results of postoperative intrathoracic chemo-thermotherapy for lung cancer with pleural dissemination. Cancer 1993;72:426-431.[Medline]
- Matsuzaki Y., Shibata K., Yoshioka M. Intrapleural perfusion hyperthermo-chemotherapy for malignant pleural dissemination and effusion. Ann Thorac Surg 1995;59:127-131.[Abstract/Free Full Text]
- Ratto G.B., Civalleri D., Esposito M., et al. Pleural space perfusion with cisplatin in the multimodality treatment of malignant mesothelioma: a feasibility and pharmacokinetic study. J Thorac Cardiovasc Surg 1999;117:759-765.[Abstract/Free Full Text]
- Yoshioka M., Shibata K., Matsuzaki Y., et al. Experimental study on acute heat injury of the lung induced by whole body hyperthermia. Jpn J Hypertherm Oncol 1986;2:181-190.
- Rafaely Y., Simansky D.A., Paley M., Gottfried M., Yellin A. Resection and perfusion thermochemotherapy: a new approach for the treatment of thymic malignancies with pleural spread. Ann Thorac Surg 2001;72:366-370.[Abstract/Free Full Text]
- Yasumoto K., Shimokawa T., Nagashima A., et al. Pharmacokinetics of cisplatin instilled into the pleural cavity following panpleuropneumonectomy in patients with malignant pleurisy due to lung cancer. J Surg Oncol 1993;54:67-70.[Medline]
- Herman T.S., Teicher B.A., Collins L.S. Effect of hyperthermia and acidosis on the cytotoxicity of four platinum complexes at normal and hyperthermic temperatures. Cancer Res 1988;48:2342-2347.[Abstract/Free Full Text]
- Zimm S., Clearly S.M., Lucas W.E., Howell S.B. Phase I/pharmacokinetic study of intraperitoneal cisplatin and etoposide. Cancer Res 1987;47:1712-1716.[Abstract/Free Full Text]
- Rusch V.W., Figlin R., Godwin D., Piantadosi S. Intrapleural cisplatin and cytarabine in the management of malignant pleural effusions: a lung cancer study group trial. J Clin Oncol 1991;9:313-319.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
M. Lucchi and A. Mussi
Reply to Heyman and Van Schil
Eur. J. Cardiothorac. Surg.,
September 1, 2008;
34(3):
708 - 708.
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. de Bree, S. van Ruth, C. E. Schotborgh, P. Baas, and F. A. N. Zoetmulder
Limited Cardiotoxicity after Extensive Thoracic Surgery and Intraoperative Hyperthermic Intrathoracic Chemotherapy with Doxorubicin and Cisplatin
Ann. Surg. Oncol.,
October 1, 2007;
14(10):
3019 - 3026.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
V. Patel and J. B. Shrager
Which Patients with Stage III Non-Small Cell Lung Cancer Should Undergo Surgical Resection?
Oncologist,
May 1, 2005;
10(5):
335 - 344.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
