Q J Med 1999; 92: 573-578
© 1999 Association of Physicians
Multiple logistic regression analysis of plasma paraquat concentrations as a predictor of outcome in 375 cases of paraquat poisoning
From the National Poisons Information Service (London), Medical Toxicology Unit, Guy's and St Thomas' Hospital Trust, London, and 1 Medical Statistics Unit, University of Edinburgh Medical School, Edinburgh, UK
Received 22 February 1999 and in revised form 6 August 1999
Dr A.L. Jones, National Poisons Information Service (London), Medical Toxicology Unit, Guy's and St Thomas' Hospital Trust, Avonley Road, London SE14 5ER
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Successful prediction of who may survive paraquat poisoning can prevent inappropriately aggressive treatment in those who have little hope of survival and those only minimally poisoned. We examined case records of patients admitted to one poisoning treatment unit over the last 5 years, and the English and French language literature on paraquat poisoning. Data were recorded from all patients where outcome and timed plasma paraquat concentrations were present. Of 375 patients (113 M, 62 F, 200 unknown), mean age 38.3 years (range 1–87 years), 49 had evidence of renal toxicity, and 41 received haemodialysis or charcoal haemoperfusion; 61 developed pulmonary sequelae; and 44 had lesions in the upper gastrointestinal tract. Median time from ingestion to death in the 241 deaths reported was 270 h (range 3–720 h). We plotted log(plasma paraquat concentration) against log(h since ingestion). The predicted probability of survival for any specified time and concentration was exp (logit)/[1+ exp (logit)], where logit=0.58–2.33xlog(plasma paraquat)–1.15xlog(h since ingestion). This equation may be helpful in predicting who will survive after ingestion of paraquat up to at least 200 h after ingestion, and can now be used as a research tool for studies on efficacy of treatment of paraquat poisoning.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Paraquat remains a major cause of death in developing countries, such as Pakistan and Sri Lanka,1 although deaths also occur in the UK each year.2 Although there are few efficacious therapeutic options for the management of paraquat poisoning, it is important to be able to predict who will survive, so that inappropriately aggressive techniques, such as haemodialysis, are not used in those who have no hope of survival, and to advise patients and relatives of the likely outcome. Similarly, minimally-poisoned patients may be protected from unnecessarily aggressive treatment.
Proudfoot3 examined 79 cases of paraquat poisoning, and related outcome to plasma paraquat concentrations on admission and the ingestion to sampling interval. This has guided therapy, at least in the UK, for the last two decades; however, the numbers in this study were relatively small, the separation between patients who died and those who survived in the first 5 h after ingestion was not clear, and no patients presenting >24 h after ingestion were included. However, some patients seek medical help only once they have developed symptoms of poisoning, which is often later than 24 h.
Our aim was to examine all the clinical records of patients under our care with paraquat poisoning, and all the cases reported in the world literature, in order to produce a more data-intensive survival curve and one that was valid beyond 24 h after ingestion.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
We examined the case records of patients admitted to the poisons ward in Edinburgh Royal Infirmary over the last 5 years, and the world literature on paraquat poisoning. Mbase and Medline were searched using the terms `intoxication or poisoning, or overdose' to link with `paraquat or herbicide or pesticide'. Only English or French language publications could be included for logistical reasons.
Data were as recorded from all papers in which outcome data and timed plasma paraquat concentrations were given. Sometimes this required reading data from photocopied enlargements of graphical information. Duplication of case data between publications was excluded by examining acknowledgements of other data sources on papers and by closely matching patient criteria (i.e. age, sex, plasma paraquat concentration, and time since ingestion) between all publications. When there was any doubt about duplication, the data was entered only once for analysis.
When serial plasma paraquat concentrations were available from a single patient, the first result available to the clinician was used. The theoretical basis for this approach comes from the observation that when multiple samples were taken for paraquat estimation, `peak plasma paraquat concentrations' had been reached before the first samples were drawn, even when the interval between ingestion and sampling was as short as 2 h.3 This is also important, as data for this study have been collected before elimination methods have been used, although in any case, their efficacy is highly questionable.4
Plasma paraquat concentrations related to time since ingestion in patients who survived and those who died were plotted using GraphPad (Prism 1996, Version 2.01). Multiple logistic regression analysis was performed to provide a line that fitted the data and predicted survival or death, using a standard statistical program (SPSS Version 4.01, 1994).
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Data were collected from papers in English or French languages.2–57 A list of those omitted from the study, either inaccessible because of language or with unavailable data, is available on request. We identified 375 cases (113 men, 62 women, and 200 of unidentifiable sex), of mean age 38.3 years (range 1–87 years). They included patients from Australia, Belgium, France, the French West Indies, Germany, Holland, Israel, Japan, South America, Taiwan, Trinidad and Tobago, the UK, and the USA. Preparations taken included 10%, 12%, 20% and 24% paraquat. In 106 cases the route of exposure was noted: one intravenous, one oral, five dermal and, 99 by mouth. Several were severely poisoned: 49 had evidence of renal toxicity, of whom 41 eventually received haemodialysis and/or charcoal haemoperfusion; 61 developed pulmonary sequelae and in six, supplemental oxygen was given; 18 patients had evidence of hepatic injury; 44 had lesions of the upper gastrointestinal tract; seven patients developed circulatory failure. The median time from ingestion to death in the 241 deaths reported was 270 h (range 3 to 720 h).
A plot of the logarithm of the plasma paraquat concentration versus the logarithm of the time since ingestion is shown in Figure 1
. A logistic regression of survival using both these variables gave significant independent effects for each (
2=95.3 for plasma concentration and 2=13.5 for time, both p<0.001).
|
To calculate a predicted probability of survival for any specified time and concentration, a formula was derived based on the logistic regression coefficients:
 |
 |
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Paraquat poisoning remains a significant world-wide cause of morbidity and mortality.1 There have been several previous attempts to predict survival in patients poisoned with paraquat using plasma paraquat concentrations and/or clinical criteria.3,38,42,46,49,59–61 By far the most useful and in widest use, despite the limitations discussed above, has been the survival curve produced by Proudfoot.3,49
Several probability survival curves based on a larger number of patients (n=218) were provided by Hart,58 but similarly, the data did not extend beyond 28 h after ingestion of paraquat. In the non-exponential mathematical survival curve produced by Scherrman,38 data were included up to 360 h after ingestion, but this curve was based on only 30 patients. Ikebuchi61 performed multivariate analysis on data from 128 poisoned patients in order to assess the severity of paraquat poisoning. As a result, the TIP or toxicological index of paraquat was developed. However, only 21 patients who survived were included in this study.
We report a new relation between plasma paraquat concentration and time, which may be helpful in predicting who will survive after ingestion of paraquat up to at least 200 h after ingestion (Figure 1
). Although a line has been drawn in Figure 1 to separate the fatal and non-fatal cases, it can only act as a guide, and prognosis may be influenced by several factors, including individual sensitivity to paraquat58 and inaccuracies in assessment of the ingestion-presentation interval.
One patient appears anomalous (Figure 1), a 52-year-old man with a past history of alcoholism and seizures who had ingested `one glass of herbicide' and who died in spite of a very low concentration of paraquat: 0.0024 mg at 13 h after ingestion.40 We could find no obvious explanation for this anomaly, unless the patient gave the time from ingestion incorrectly or a laboratory error occurred.40 Despite intensive supportive care and treatment with fuller's earth and haemodialysis, he died as a result of cardiovascular shock and acute renal failure, having developed the characteristic corrosive gastrointestinal tract lesions of paraquat toxicity.40 Thus, the data on this patient have been included in the analysis.
Use of the survival curve (Figure 1) is not intended to replace the initial qualitative screening test, which is performed by adding a knifepoint of sodium dithionite and a knifepoint of sodium bicarbonate to a urine sample taken within 24 h of ingestion of paraquat.38,62 However, if the patient presents >24 h after ingestion, the qualitative urine test is not reliable and should not be used.38 In that event, plasma paraquat concentrations may be used, together with Figure 1, to help predict survival.
The logistic regression using both time and plasma paraquat concentration as variables gave very highly significant independent effects for each as predictors of survival. This quantifies what is apparent from Figure 1, namely that patients with a given concentration of paraquat have a better chance of survival if they have only recently taken it, and at any time after ingestion, the prognosis is better the lower the plasma concentration. Use of the logistic regression equation allows us to predict the probability of survival for any specified time and paraquat concentration. Whilst calculation looks quite complicated, it is remarkably easy to do on a pocket calculator. It could be used to draw in contours of constant probability on the plot, and these would be straight lines parallel to the one already shown in Figure 1. For example, a contour for a 50% chance of survival would correspond to a logit of zero, giving a line defined by:
 |
Those patients near the survival line (Figure 1) might be anticipated to be those most likely to benefit from therapeutic intervention; a prospective multicentre study should evaluate whether measures which have been previously used for gut decontamination, such as activated charcoal12,63,64 or for elimination14,22,30,37 will alter outcome. There are many reports of the ineffectiveness of such strategies, but the patients have not been sufficiently standardized according to risk to allow valid comparisons between studies or groups of patients. As the absorption of paraquat peaks at 2 h after ingestion, and irreversible fixation of paraquat into the alveolar cells occurs within the first 4 h, any technique designed to increase elimination of absorbed paraquat must be instituted as early as possible in order to remove toxicologically significant quantities,64 and most elimination methods, such as haemoperfusion and haemodialysis, appear to remove only a very small proportion of absorbed paraquat.65 When renal function is conserved, elimination by the kidney is 3–10 times more efficient than haemoperfusion.65 It is therefore possible that methods for enhancing elimination may prove ineffective.
Reduction of morbidity and mortality of poisoning by paraquat also relies on methods designed to prevent ingestion of significant amounts of toxin. This includes limiting the supply of concentrates, the addition of stenching and emetic agents, and adequate labelling and public awareness of the hazard.63
In conclusion, prognosis in acute paraquat poisoning is largely determined by the time between ingestion and plasma paraquat concentrations before treatment. This curve that we have described (Figure 1) and the logistic equation associated with it have extended the original Proudfoot data3 to later times and have provided a relatively simple mathematical way of calculating the risk of death. This can now be used as a research tool for studies on the efficacy of treatment of paraquat poisoning. It is also important that the new survival curve is now validated prospectively, to determine its specificity and sensitivity for predicting outcome in patients poisoned with paraquat.
|
|
Acknowledgments |
|---|
We would like to thank Dr A.T. Proudfoot, the former Director of the Scottish Poisons Information Bureau for permission to include data on five patients under his care at Edinburgh Royal Infirmary, and his advice and constructive criticism regarding this work.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Jeyaratnam J. Acute pesticide poisoning: a major global health problem. World Health Stat Q 1990; 43:139–44.[Medline]
2.
Thompson JP, Casey PB, Vale JA. Deaths from pesticide poisoning in England and Wales 1990–1991. Hum Exp Toxicol 1995; 14:437–45.
3. Proudfoot AT, Stewart MS, Levitt T, Widdop, B. Paraquat poisoning: significance of plasma-paraquat concentrations. Lancet 1979; 2:330–2.[Web of Science][Medline]
4. Van de Vyver FL, Giuliano RA, Paulus GJ, Verpooten GA, Franke JP, de Zeeuw RA, van Gaal LF, De Broe ME. Hemoperfusion-hemodialysis ineffective for paraquat removal in life-threatening poisoning. J Toxicol Clin Toxicol 1985; 23:117–31.[Web of Science][Medline]
5.
Hargreave TB, Gresham GA, Karayannopoulos S. Paraquat poisoning. Postgrad Med J 1969; 45:633–5.
6. Malone JDG, Carmody M, Keogh B, O'Dwyer WF. Paraquat poisoning—a review of nineteen cases. J Irish Med Assoc 1971; 64:59–68.[Medline]
7. Davidson JK, MacPherson P. Pulmonary changes in paraquat poisoning. Clin Radiol 1972; 23:18–25.[Medline]
8. Douglas JF, McGeown MG, McEvoy J. The treatment of paraquat poisoning: three cases of recovery. Ulster Med J 1973; 42:209–12.[Medline]
9. Eliahou HE, Almog C, Gura V, Iaina A. Treatment of paraquat poisoning by hemodialysis. Isr J Med Sci 1973; 9:459–62.[Web of Science][Medline]
10. Okonek S, Hofmann A, Henningsen B. Efficacy of gut lavage, hemodialysis, and hemoperfusion in the therapy of paraquat or diquat intoxication. Arch Toxicol 1976; 36:43–51.[Web of Science][Medline]
11. Sofrank G, Mathes GV, Clarmann M, Beyer KH. Haemoperfusion through activated charcoal in paraquat intoxication. Acta Pharmacol Toxicol (Copenh) 1977; 41:91–101.
12. Vale JA, Crome P, Volans GN, Widdop B, Goulding R. The treatment of paraquat poisoning using oral sorbents and charcoal haemoperfusion. Acta Pharmacol Toxicol (Copenh) 1977; 41:109–17.
13. Ackrill P, Hasleton PS, Ralston AJ. Oesophageal perforation due to paraquat. Br Med J 1978; 1:1252–3.
14. Dearnaley DP, Martin MFR. Plasmapheresis for paraquat poisoning. Lancet 1978; 1:162.
15. Spector D, Whorton D, Zachary J, Slavin R. Fatal paraquat poisoning: tissue concentrations and implications for treatment. Johns Hopkins Med J 1978; 142:110–13.[Web of Science][Medline]
16. Fairshter RD, Dabir-Vaziri N, Smith WR, Glauser FL, Wilson AF. Paraquat poisoning: an analytical toxicologic study of three cases. Toxicology 1979; 12:259–66.[Web of Science][Medline]
17.
Higenbottam T, Crome P, Parkinson C, Nunn J. Further clinical observations on the pulmonary effects of paraquat ingestion. Thorax 1979; 34:161–5.
18. Rose JD. Paraquat poisoning (letter). Lancet 1980; 2:924.
19. Russell LA, Stone BE, Rooney PA. Paraquat poisoning: toxicologic and pathologic findings in three fatal cases. Clin Toxicol 1981; 18:915–28.[Web of Science][Medline]
20. Vincken W, Huyghens L, Schandevyl W, Verbeelen D, Corne L. Paraquat poisoning and colchicine treatment (letter). Ann Intern Med 1981; 95:391–2.
21. Bismuth C, Garnier R, Dally S, Fournier PE. Prognosis and treatment of paraquat poisoning: a review of 28 cases. J Toxicol Clin Toxicol 1982; 19:461–74.[Web of Science][Medline]
22. De Groot G. Haemoperfusion in paraquat intoxication. In: Haemoperfusion in Clinical Toxicology: A Pharmacokinetic Evaluation, Vol. 1. Utrecht, These, 1982.
23. Mofenson HC, Greensher J, Caraccio TR, Agostino R. Paraquat intoxication: report of a fatal case. Discussion of pathophysiology and rational treatment. J Toxicol Clin Toxicol 1982; 19:821–34.[Medline]
24.
Musson FA, Porter CA. Effect of ingestion of paraquat on a 20-week gestation fetus. Postgrad Med J 1982; 58:731–2.
25. Siefkin AD. Combined paraquat and acetaminophen toxicity. J Toxicol Clin Toxicol 1982; 19:483–91.[Web of Science][Medline]
26. Garnier R, Efthymiou ML, Baud F. Haemoperfusion for paraquat poisoning. Lancet 1983; ii:277.
27.
Hoffman S, Jedeikin R, Korzets Z, Shapiro A, Kaplan R, Bernheim J. Successful management of severe paraquat poisoning. Chest 1983; 84:107–9.
28. Mascie-Taylor BH, Thompson J, Davison AM. Haemoperfusion ineffective for paraquat removal in life-threatening poisoning. Lancet 1983; 1:1376–7.[Medline]
29. Scherrman JM, Galliot M, Garnier R, Bismuth C. Acute paraquat poisoning: prognostic significance and therapeutic interest of blood assay. Toxicol Eur Res 1983; 5:141–5.[Medline]
30. Van de Vyver FL, van de Sande J, Verpooten GA, De Broe ME, van den Heede M, Heyndrickx A. Haemoperfusion ineffective for paraquat removal in life-threatening poisoning. Lancet 1983; 2:1376–7.
31.
Hendy MS, Williams PS, Ackrill P. Recovery from severe pulmonary damage due to paraquat administered intravenously and orally. Thorax 1984; 39:874–5.
32. Webb DB, Williams MV, Davies BH, James KW. Resolution after radiotherapy of severe pulmonary damage due to paraquat poisoning. Br Med J 1984; 288:1259.
33. Addo E, Poon King T. Leucocyte suppression in treatment of 72 patients with paraquat poisoning. Lancet 1986; 1:1117–20.[Web of Science][Medline]
34. Bismuth C, Baud F, Dally S. Standardised prognosis evaluation in acute toxicology its benefit in colchicine, paraquat and digitalis poisonings. J Toxicol Clin Exp 1986; 6:33–8.[Medline]
35.
Yasaka T, Okufddaira K, Fujito H, Atsumoto J, Ohya I, Miyamoto Y. Further studies of lipid peroxidation in human paraquat poisoning. Arch Intern Med 1986; 146:681–5.
36. Ikebuchi J. Evaluation of paraquat concentrations in paraquat poisoning. Arch Toxicol 1987; 60:304–10.[Web of Science][Medline]
37. Pond SM, Johnston SC, Schoof DD, Hampson EC, Bowles M, Wright DM, Petrie JJ. Repeated hemoperfusion and continuous arteriovenous hemofiltration in a paraquat poisoned patient. J Toxicol Clin Toxicol 1987; 25:305–16.[Medline]
38. Scherrman JM, Houze P, Bismuth C, Bourdon R. Prognostic value of plasma and urine paraquat concentration. Hum Toxicol 1987; 6:91–3.[Web of Science][Medline]
39. Williams MV, Webb DB. Paraquat lung is there a role for radiotherapy? Hum Toxicol 1987; 6:75–81.[Web of Science][Medline]
40. Baud FJ, Houze P, Bismuth C, Scherrman J, Jaeger A, Keyes C. Toxicokinetics of paraquat through the heart-lung block. Six cases of acute human poisoning. J Toxicol Clin Toxicol 1988; 26:35–50.[Medline]
41. Hampson ECGM, Pond SM. Failure of haemoperfusion and haemodialysis to prevent death in paraquat poisoning. A retrospective review of 42 patients. Med Toxicol 1988; 3:64–71.[Web of Science][Medline]
42. Sawada Y, Yamamoto I, Hirokane T, Nagai Y, Satoh Y, Ueyama M. Severity index of paraquat poisoning. Lancet 1988; 1:1333.
43. Talbot AR, Barnes MR. Radiotherapy for the treatment of pulmonary complications of paraquat poisoning. Hum Toxicol 1988; 7:325–32.[Web of Science][Medline]
44. Talbot AR, Fu CC, Hsieh MF. Paraquat intoxication during pregnancy: a report of 9 cases. Vet Hum Toxicol 1988; 30:12–17.
45. Hoffer E, Taitelman U. Exposure to paraquat through skin absorption: clinical and laboratory observations of accidental splashing on healthy skin of agricultural workers. Hum Toxicol 1989; 8:483–5.[Medline]
46. Suzuki K, Takasu N, Arita S, Maenosono A, Ishimatsu S, Nishina M, Tanaka S, Kohama A. A new method for predicting the outcome and survival period in paraquat poisoning. Hum Toxicol 1989; 8:33–8.[Web of Science][Medline]
47.
Franzen D, Baer F, Heitz W, Mecking H, Eidt S, Käferstein H, Baldamus CA, Curtius JM, Höpp HW, Wassermann K. Failure of radiotherapy to resolve fatal lung damage due to paraquat poisoning. Chest 1991; 100:1164–5.
48. Ragoucy-Sengler C, Pileire B, Daijardin JB. Survival from severe paraquat intoxication in heavy drinkers. Lancet 1991; 338:1461.[Web of Science][Medline]
49.
Suzuki K, Nobukatsu T, Arita S, Ueda A, Okabe T, Ishimatsu S, Tanaka S, Kohama A. Evaluation of severity indexes of patients with paraquat poisoning. Hum Exp Toxicol 1991; 10:21–3.
50.
Florkowski CM, Bradberry SM, Ching GWK, Jones AF. Acute renal failure in a case of paraquat poisoning with relative absence of pulmonary toxicity. Postgrad Med J 1992; 68:660–2.
51. Perriens JH, Benimadho S, Kiauw IL, Wisse J, Chee H. High-dose cyclophosphamide and dexamethasone in paraquat poisoning: a prospective study. Hum Exp Toxicol 1992; 11:129–34.[Web of Science][Medline]
52. Rivero C, Martinez R, Gonzalez E, Espinoza OB, Ramirez MS. Paraquat poisoning in children: survival of three cases. Vet Hum Toxicol 1992; 34:164–5.[Medline]
53. Suzuki K, Takasu N, Okabe T, Ishimatsu S, Ueda A, Tanaka S, Fukuda A, Arita S, Kohama A. Effect of aggressive haemoperfusion on the clinical course of patients with paraquat poisoning. Hum Exp Toxicol 1993; 12:323–7.[Web of Science][Medline]
54. Casey PB, Buckley BM, Vale JA. Methemoglobinemia following ingestion of a monolinuron/paraquat herbicide (Gramanol). Clin Toxicol 1994; 32:185–9.
55. Garnier R, Chataigner D, Efthymiou M, Moraillon I, Bramary F. Paraquat poisoning by skin absorption: report of two cases. Vet Hum Toxicol 1994; 36:313–15.[Medline]
56. Koppel C, Wissmann CV, Barckow D, Rossaint R, Falke K, Stoltenburg-Didinger G, Schnoy N. Inhaled nitric oxide in advanced paraquat intoxication. Clin Toxicol 1994; 32:205–14.[Medline]
57.
Lheureux P, Leduc D, Vanbinst R, Askenasi R. Survival in a case of massive paraquat ingestion. Chest 1995; 107:285–9.
58. Hart TB, Nevitt A, Whitehead A. A new statistical approach to the prognostic significance of plasma paraquat concentration. Lancet 1984; 1:1222.[Web of Science][Medline]
59. Yamaguchi H, Sato S, Watanabe S, Naito H. Pre-embarkment prognostication for acute paraquat poisoning. Hum Exp Toxicol 1990; 9:381–4.[Web of Science][Medline]
60. Kaojarern S, Ongphiphadhanakul B. Predicting outcomes in paraquat poisonings. Vet Hum Toxicol 1991; 33:115–18.[Web of Science][Medline]
61. Ikebuchi J, Proudfoot AT, Matsubara K, Hampson ECGM, Tomita M, Suzuki K, Fuke C, Ijiri I, Tsunerari T, Yuasa I, Okada K. Toxicological index of paraquat: a new strategy for assessment of severity of paraquat poisoning in 128 patients. Forens Sci Int 1993; 59:85–7.
62. Berry DJ, Grove J. The determination of paraquat cation in urine. Clin Chim Acta 1971; 34:5–11.[Web of Science][Medline]
63. Meredith TJ, Vale JA. Treatment of paraquat poisoning in man Methods to prevent absorption. Hum Toxicol 1987; 6:49–55.[Web of Science][Medline]
64. Honoré P, Hantson P, Fauville JP, Peeters A, Manieu P. Paraquat poisoning. `State of the art'. Acta Clin Belg 1994; 49:220–8.[Medline]
65. Proudfoot AT, Prescott LF, Jarvie DR. Haemodialysis for paraquat poisoning. Hum Toxicol 1987; 6:69–74.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Senarathna, M. Eddleston, M.F. Wilks, B.H. Woollen, J.A. Tomenson, D.M. Roberts, and N.A. Buckley Prediction of outcome after paraquat poisoning by measurement of the plasma paraquat concentration QJM, April 1, 2009; 102(4): 251 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Buckley and J.-L. Lin PULSE CORTICOSTEROID AND CYCLOPHOSPHAMIDE IN PARAQUAT POISONING Am. J. Respir. Crit. Care Med., February 1, 2001; 163(2): 585 - 585. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||